[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"portal-settings:stajic:sr":3,"public-menus:all":38,"post:source-of-truth-in-ai-systems-where-reliable-knowledge-actually-comes-from:sr":205,"related:post:source-of-truth-in-ai-systems-where-reliable-knowledge-actually-comes-from:sr:1":2782},{"statusCode":4,"data":5,"message":37},200,{"tenantId":6,"lang":7,"defaultLang":8,"siteUrl":9,"contactEmail":10,"brandName":11,"logoUrl":12,"siteName":11,"siteDescription":13,"ogImage":10,"robotsIndex":14,"socialLinks":10,"reservedSlugs":10,"seoPolicy":15},"stajic","sr","de","https:\u002F\u002Fstajic.de",null,"Stajic Platform","\u002FLogo_Planet.svg","Stajic Portal",true,{"branding":16,"relatedContent":17,"crossDomainLinks":18},{"logoUrl":12},{"enabled":14},[19,22,25,28,31,34],{"url":20,"label":21,"isActive":14,"showInFooter":14,"includeInSameAs":14},"https:\u002F\u002Ffigure.rocks","figure.rocks",{"url":23,"label":24,"isActive":14,"showInFooter":14,"includeInSameAs":14},"https:\u002F\u002Floving.rocks","loving.rocks",{"url":26,"label":27,"isActive":14,"showInFooter":14,"includeInSameAs":14},"https:\u002F\u002Fbazify.com","bazify.com",{"url":29,"label":30,"isActive":14,"showInFooter":14,"includeInSameAs":14},"https:\u002F\u002Fbazify.de","bazify.de",{"url":32,"label":33,"isActive":14,"showInFooter":14,"includeInSameAs":14},"https:\u002F\u002Fbazify.at","bazify.at",{"url":35,"label":36,"isActive":14,"showInFooter":14,"includeInSameAs":14},"https:\u002F\u002Fbazify.ba","bazify.ba","Portal settings resolved",[39,45],{"id":40,"name":41,"location":42,"isActive":14,"isDefault":43,"items":44},1,"main-navigation","header",false,[],{"id":46,"name":47,"location":48,"isActive":14,"isDefault":14,"items":49},4,"main-menu","sidebar",[50,66,79,93,103,118,133],{"id":51,"title":52,"url":60,"target":61,"icon":62,"isActive":14,"type":63,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":64,"portfolioId":10,"children":65},"item-18",{"de":53,"en":54,"es":55,"fr":56,"it":54,"ru":57,"sr":58,"zh":59},"Startseite","Home","Inicio","Accueil","Главная","Почетна","首页","\u002Ffull-stack-web-developer-munich-performance-seo-and-maintainable-builds","_self","i-lucide-home","page",111,[],{"id":67,"title":68,"url":75,"target":61,"icon":76,"isActive":14,"type":63,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":77,"portfolioId":10,"children":78},"item-22",{"de":69,"en":69,"es":70,"fr":69,"it":71,"ru":72,"sr":73,"zh":74},"Vision","Visión","Visione","Видение","Визија","想象","\u002Fueber-uns-webdesign-muenchen-webaplikation","i-lucide-eye",113,[],{"id":80,"title":81,"url":89,"target":61,"icon":90,"isActive":14,"type":63,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":91,"portfolioId":10,"children":92},"item-19",{"de":82,"en":83,"es":84,"fr":83,"it":85,"ru":86,"sr":87,"zh":88},"Leistungen","Services","Servicios","Servizi","Услуги","Услуге","服务","\u002Fservices-dienstleistungen-muenchen","i-lucide-wrench",116,[],{"id":94,"title":95,"url":99,"target":61,"icon":100,"isActive":14,"type":63,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":101,"portfolioId":10,"children":102},"item-23",{"de":96,"en":96,"es":96,"fr":96,"it":96,"ru":97,"sr":97,"zh":98},"Blog","Блог","博客","\u002Fblog","i-lucide-book-open",112,[],{"id":104,"title":105,"url":114,"target":61,"icon":115,"isActive":14,"type":63,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":116,"portfolioId":10,"children":117},"item-32",{"de":106,"en":107,"es":108,"fr":109,"it":110,"ru":111,"sr":112,"zh":113},"Neue Technologien","New Technologies","Nuevas tecnologías","Nouvelles technologies","Nuove tecnologie","Новые технологии","Нове технологије","新技术！","\u002Fneue-webtechnologien","i-lucide-sparkles",122,[],{"id":119,"title":120,"url":129,"target":61,"icon":130,"isActive":14,"type":63,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":131,"portfolioId":10,"children":132},"item-20",{"de":121,"en":122,"es":123,"fr":124,"it":125,"ru":126,"sr":127,"zh":128},"Kontakt","Contact us!","Contacto","Contact","Contatto","Контакт","Контактирајте нас","联系我们！","\u002Fcontact","i-lucide-mail",115,[],{"id":134,"title":135,"url":144,"target":61,"icon":145,"isActive":14,"type":63,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":146,"portfolioId":10,"children":147},"item-21",{"de":136,"en":137,"es":138,"fr":139,"it":140,"ru":141,"sr":142,"zh":143},"Unsere Arbeit","Our Work","Nuestro trabajo","Nos réalisations","I nostri lavori","Наши работы","Наши радови","文件夹","\u002Fportfolio","i-lucide-briefcase",114,[148,161,175,181,193],{"id":149,"title":150,"url":144,"target":61,"icon":159,"isActive":14,"type":63,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":146,"portfolioId":10,"children":160},"item-24",{"de":151,"en":152,"es":153,"fr":154,"it":155,"ru":156,"sr":157,"zh":158},"Alle Projekte","All Projects","Todos los proyectos","Tous les projets","Tutti i progetti","Все проекты","Сви пројекти","所有项目","i-lucide-grid-3x3",[],{"id":162,"title":163,"url":171,"target":61,"icon":172,"isActive":14,"type":173,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":10,"portfolioId":10,"children":174},"item-29",{"de":164,"en":165,"es":166,"fr":167,"it":168,"ru":169,"sr":170,"zh":143},"Local Roots, Global Reach","Local Roots - Global Reach","Empresa local ","Entreprise locale","Azienda locale","Местная компания","Локално предузеће глобално тржиште","\u002Fportfolio\u002Flocal-roots-global-reach-communication-media-systems-for-modern-business","i-lucide-folder","custom",[],{"id":176,"title":177,"url":179,"target":61,"icon":172,"isActive":14,"type":173,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":10,"portfolioId":10,"children":180},"item-28",{"de":178,"en":178,"es":178,"fr":178,"it":178,"ru":178,"sr":178,"zh":178},"Solr Suggester","\u002Fportfolio\u002Fsolr-fuzzy-suggester-und-solr-infix-suggester-abfrage-ueber-ajax-und-filterung",[],{"id":182,"title":183,"url":191,"target":61,"icon":172,"isActive":14,"type":173,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":10,"portfolioId":10,"children":192},"item-27",{"de":184,"en":185,"es":186,"fr":187,"it":188,"ru":189,"sr":190,"zh":185},"Firmenwebseite SEO","Company Website SEO","Sitio web corporativo SEO","Site web d’entreprise SEO","Sito web aziendale SEO","Корпоративный сайт SEO","Пословна веб-страница SEO","\u002Fportfolio\u002Fseo-sem-branding-mobile-webseite-muenchen",[],{"id":194,"title":195,"url":203,"target":61,"icon":172,"isActive":14,"type":173,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":10,"portfolioId":10,"children":204},"item-31",{"de":196,"en":197,"es":198,"fr":199,"it":200,"ru":201,"sr":202,"zh":197},"Digitalisierungsportal","Digitalization Portal","Portal de digitalización","Portail de numérisation","Portale di digitalizzazione","Портал цифровизации","Портал за дигитализацију","\u002Fportfolio\u002Fdigitalisierungsportal-archiv-museum-bibliothek-ead-lido-mets-mods",[],{"statusCode":4,"data":206,"message":2781},{"id":207,"title":208,"slug":209,"content":210,"contentJson":211,"excerpt":1297,"featuredImage":1298,"featuredImageAlt":1299,"featuredImageCaption":10,"featuredImageTitle":10,"featuredImageCopyright":10,"featuredImageAuthor":10,"featuredImageSourceUrl":10,"featuredImageLicense":10,"featuredImageIsAiGenerated":43,"status":1300,"publishedAt":1301,"createdAt":1302,"updatedAt":1303,"seoLocalePaths":1304,"categories":1313,"author":1326,"translations":1331},"486","Izvor istine u AI sistemima: Odakle pouzdano znanje zaista dolazi","source-of-truth-in-ai-systems-where-reliable-knowledge-actually-comes-from","\u003Cp>Izvor istine u AI sistemu je autoritativni izvor koji je ovlašćen da definiše da li određena činjenica, stanje ili pravilo treba da se tretiraju kao istiniti za određeni obim, verziju i vreme. To automatski nije jezički model, vektorska baza podataka, najbolje rangirani preuzeti dokument, memorija agenta ili najnovija poruka u kontekstu. Pouzdana AI arhitektura mora da očuva koji izvor ima autoritet za koju tvrdnju, a zatim da održi provenijenciju, preuzimanje i validaciju povezane sa tim autoritetom.\u003C\u002Fp>\n\u003Caside class=\"editorjs-callout editorjs-callout--info my-6 rounded-xl border p-5 border-blue-300 bg-blue-50 dark:border-blue-900 dark:bg-blue-950\u002F20\" role=\"note\">\u003Cstrong class=\"block mb-2 text-gray-900 dark:text-gray-100\">Direktan odgovor\u003C\u002Fstrong>\u003Cdiv class=\"text-gray-700 dark:text-gray-200\">\u003Cstrong>Izvor istine je pravilo autoriteta, a ne AI komponenta.\u003C\u002Fstrong> AI sistem može da pretražuje mnoge izvore, ali samo neki izvori su autoritativni za dato pitanje. Ispravan izvor može biti aplikaciona baza podataka za trenutno stanje naloga, odobreni dokument politike za pravilo, izvorni kod za implementirano ponašanje, zvanična specifikacija za protokol ili primarni istraživački artefakt za uočenu tvrdnju. Dobra arhitektura čini taj autoritet eksplicitnim umesto da traži od modela da ga zaključi na osnovu relevantnosti.\u003C\u002Fdiv>\u003C\u002Faside>\n\u003Caside class=\"editorjs-callout editorjs-callout--note my-6 rounded-xl border p-5 border-gray-300 bg-gray-50 dark:border-gray-700 dark:bg-gray-900\u002F40\" role=\"note\">\u003Cstrong class=\"block mb-2 text-gray-900 dark:text-gray-100\">Terminološka granica\u003C\u002Fstrong>\u003Cdiv class=\"text-gray-700 dark:text-gray-200\">„Izvor istine“ se široko koristi u softveru, podacima i organizacionoj praksi, ali nije standardizovana AI komponenta sa jednom univerzalnom formalnom definicijom. Ovaj članak ga koristi kao arhitektonski koncept: \u003Cstrong>izvor ili mapiranje autoriteta koje određuje koji dokazi smeju da utvrde određenu činjenicu ili stanje\u003C\u002Fstrong>. Standardi provenijencije kao što je W3C PROV opisuju poreklo i izvođenje; oni automatski ne odlučuju koji izvor vaša aplikacija treba da tretira kao autoritativan.\u003C\u002Fdiv>\u003C\u002Faside>\n\u003Caside class=\"editorjs-callout editorjs-callout--note my-6 rounded-xl border p-5 border-gray-300 bg-gray-50 dark:border-gray-700 dark:bg-gray-900\u002F40\" role=\"note\">\u003Cstrong class=\"block mb-2 text-gray-900 dark:text-gray-100\">Napomena o trenutnom izvoru — 8. oktobar 2026.\u003C\u002Fstrong>\u003Cdiv class=\"text-gray-700 dark:text-gray-200\">Osnovna razlika između autoriteta, provenijencije, preuzimanja i konteksta modela je stabilna. NIST AI RMF 1.0 se trenutno revidira, ali trenutni NIST Playbook i dalje eksplicitno preporučuje dokumentovanje provenijencije podataka, uključujući izvore, poreklo, transformacije, zavisnosti, ograničenja i metapodatke. W3C PROV ostaje W3C preporuka za interoperabilne opise provenijencije.\u003C\u002Fdiv>\u003C\u002Faside>\n\u003Cnav class=\"editorjs-toc\" data-editorjs-toc=\"true\" aria-label=\"Sadržaj\">\u003Cstrong class=\"editorjs-toc__title\">Sadržaj\u003C\u002Fstrong>\u003Col class=\"editorjs-toc__list editorjs-toc__list--depth-0\">\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-6\" class=\"editorjs-toc__link\">Šta „Izvor istine“ zaista znači\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-11\" class=\"editorjs-toc__link\">Najjednostavniji primer\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-16\" class=\"editorjs-toc__link\">Gde se jednostavan primer zaustavlja\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-19\" class=\"editorjs-toc__link\">Nadležnost je ograničena tvrdnjom, verzijom i vremenom\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-22\" class=\"editorjs-toc__link\">Šta Izvor istine nije\u003C\u002Fa>\u003Col class=\"editorjs-toc__list editorjs-toc__list--depth-1\">\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-23\" class=\"editorjs-toc__link\">Izvor istine naspram sistema evidencije\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-26\" class=\"editorjs-toc__link\">Izvor istine naspram provenijencije\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-29\" class=\"editorjs-toc__link\">Izvor istine naspram dokaza\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-32\" class=\"editorjs-toc__link\">Izvor istine naspram RAG-a\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-35\" class=\"editorjs-toc__link\">Izvor istine naspram vektorske baze podataka\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-38\" class=\"editorjs-toc__link\">Izvor istine naspram memorije\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-41\" class=\"editorjs-toc__link\">Izvor istine naspram konteksta\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-44\" class=\"editorjs-toc__link\">Izvor istine naspram evaluacione osnovne istine\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-47\" class=\"editorjs-toc__link\">Izvor istine naspram kvaliteta podataka\u003C\u002Fa>\u003C\u002Fli>\u003C\u002Fol>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-50\" class=\"editorjs-toc__link\">Praktičan arhitektonski model izvora istine\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-53\" class=\"editorjs-toc__link\">Autoritet treba da bude eksplicitan, a ne izveden iz sličnosti\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-56\" class=\"editorjs-toc__link\">Pretraga treba da koristi autoritet kao ograničenje za rangiranje\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-60\" class=\"editorjs-toc__link\">Svježina je dio autoriteta\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-63\" class=\"editorjs-toc__link\">Izvedene vrijednosti zahtijevaju trag natrag do autoritativnih ulaza\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-67\" class=\"editorjs-toc__link\">Šta se dešava kada se autoritativni izvori ne slažu?\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-71\" class=\"editorjs-toc__link\">Web pretraga je otkrivanje, ne automatski dokaz\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-74\" class=\"editorjs-toc__link\">Jezički model ne treba sam da odlučuje o autoritetu\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-77\" class=\"editorjs-toc__link\">Autoritet mora da preživi trag izvršavanja\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-80\" class=\"editorjs-toc__link\">Dokaz originalne implementacije: Source of Truth Research Engine\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-87\" class=\"editorjs-toc__link\">Aaasaasa Document & Knowledge Engine: primena iste granice na poslovne dokumente\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-90\" class=\"editorjs-toc__link\">Uobičajeni načini neuspeha izvora istine\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-92\" class=\"editorjs-toc__link\">Praktični okvir za odlučivanje o izvoru istine\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-94\" class=\"editorjs-toc__link\">Kontrolna lista arhitekture izvora istine\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-96\" class=\"editorjs-toc__link\">Uobičajene zablude\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-98\" class=\"editorjs-toc__link\">Granični slučajevi\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-103\" class=\"editorjs-toc__link\">Ograničenja\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-107\" class=\"editorjs-toc__link\">Šta bi promenilo ovaj odgovor?\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-110\" class=\"editorjs-toc__link\">Povezano kanonsko znanje\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-117\" class=\"editorjs-toc__link\">Često postavljana pitanja\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-119\" class=\"editorjs-toc__link\">Pojmovnik\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-121\" class=\"editorjs-toc__link\">Zaključak\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-125\" class=\"editorjs-toc__link\">Primarni izvori i dokazi o implementaciji\u003C\u002Fa>\u003C\u002Fli>\u003C\u002Fol>\u003C\u002Fnav>\n\u003Ch2 id=\"section-6\">Šta „Izvor istine“ zaista znači\u003C\u002Fh2>\n\u003Cp>Fraza se često pogrešno razume kao „jedna baza podataka koja sadrži sve.“ To može biti tačno u uskom sistemu, ali je obično previše pojednostavljeno za AI. Prava AI aplikacija može kombinovati operativne baze podataka, dokumente, API-je, vektorske indekse, korisnički unos, memoriju modela, eksterne web izvore i generisane sažetke.\u003C\u002Fp>\n\u003Cp>Ti izvori nemaju jednak autoritet. Priručnik za korisničku podršku može definisati politiku, ali ne trenutno stanje korisnika. CRM može definisati trenutnog vlasnika naloga, ali ne pravno značenje propisa. Repozitorijum izvornog koda može definisati implementirano ponašanje, dok specifikacija proizvoda definiše nameravano ponašanje. Arhitektura stoga mora odgovoriti na preciznije pitanje: koji izvor je autoritativan za ovu konkretnu tvrdnju?\u003C\u002Fp>\n\u003Cp>Ovo čini Izvor istine odnosom između tvrdnje i autoriteta, a ne samo svojstvom tehnologije skladištenja.\u003C\u002Fp>\n\u003Caside class=\"editorjs-callout editorjs-callout--success my-6 rounded-xl border p-5 border-emerald-300 bg-emerald-50 dark:border-emerald-900 dark:bg-emerald-950\u002F20\" role=\"note\">\u003Cstrong class=\"block mb-2 text-gray-900 dark:text-gray-100\">Osnovni princip\u003C\u002Fstrong>\u003Cdiv class=\"text-gray-700 dark:text-gray-200\">\u003Cstrong>Jedan sistem može imati mnogo Izvora istine jer je autoritet specifičan za činjenicu.\u003C\u002Fstrong> Važno arhitektonsko svojstvo nije da jedna baza podataka globalno pobeđuje, već da svaka važna činjenica ili stanje ima eksplicitnog autoritativnog vlasnika.\u003C\u002Fdiv>\u003C\u002Faside>\n\u003Ch2 id=\"section-11\">Najjednostavniji primer\u003C\u002Fh2>\n\u003Cp>Korisnik pita AI asistenta: „Koji je moj trenutni plan pretplate?“ Asistent ima tri moguća ulaza: transkript podrške od prošlog meseca, indeksirani dokument centra za pomoć koji opisuje tipove planova i bazu podataka za naplatu uživo.\u003C\u002Fp>\n\u003Cp>Transkript podrške može pomenuti da je korisnik imao Pro plan. Dokument centra za pomoć objašnjava šta Pro znači. Ali zapis o naplati uživo je autoritativni izvor za trenutno stanje korisnikove pretplate.\u003C\u002Fp>\n\u003Cp>Semantički pretraživač bi mogao rangirati transkript podrške iznad zapisa o naplati jer sadrži jezik bliži pitanju. To rangiranje ipak ne bi učinilo transkript autoritativnim. Relevantnost i autoritet su različite dimenzije.\u003C\u002Fp>\n\u003Csection class=\"editorjs-comparison my-6\">\u003Ch3 class=\"mb-3 text-lg font-semibold\">Isto pitanje može uključivati različite uloge izvora\u003C\u002Fh3>\u003Cdiv class=\"overflow-x-auto\">\u003Ctable class=\"w-full border-collapse\">\u003Cthead>\u003Ctr>\u003Cth class=\"border border-gray-300 bg-gray-50 px-4 py-3 text-left dark:border-gray-700 dark:bg-gray-900\">\u003C\u002Fth>\u003Cth class=\"border border-gray-300 bg-gray-50 px-4 py-3 text-left font-semibold dark:border-gray-700 dark:bg-gray-900\">Izvor\u003C\u002Fth>\u003Cth class=\"border border-gray-300 bg-gray-50 px-4 py-3 text-left font-semibold dark:border-gray-700 dark:bg-gray-900\">Uloga\u003C\u002Fth>\u003Cth class=\"border border-gray-300 bg-gray-50 px-4 py-3 text-left font-semibold dark:border-gray-700 dark:bg-gray-900\">Autoritet za trenutni plan?\u003C\u002Fth>\u003C\u002Ftr>\u003C\u002Fthead>\u003Ctbody>\u003Ctr>\u003Cth class=\"border border-gray-300 px-4 py-3 text-left font-semibold dark:border-gray-700\">Baza podataka za naplatu\u003C\u002Fth>\u003Ctd class=\"border border-gray-300 px-4 py-3 dark:border-gray-700\">\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-3 dark:border-gray-700\">\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-3 dark:border-gray-700\">\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Cth class=\"border border-gray-300 px-4 py-3 text-left font-semibold dark:border-gray-700\">Dokumentacija centra za pomoć\u003C\u002Fth>\u003Ctd class=\"border border-gray-300 px-4 py-3 dark:border-gray-700\">\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-3 dark:border-gray-700\">\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-3 dark:border-gray-700\">\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Cth class=\"border border-gray-300 px-4 py-3 text-left font-semibold dark:border-gray-700\">Stari transkript podrške\u003C\u002Fth>\u003Ctd class=\"border border-gray-300 px-4 py-3 dark:border-gray-700\">\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-3 dark:border-gray-700\">\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-3 dark:border-gray-700\">\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Cth class=\"border border-gray-300 px-4 py-3 text-left font-semibold dark:border-gray-700\">Memorija modela\u003C\u002Fth>\u003Ctd class=\"border border-gray-300 px-4 py-3 dark:border-gray-700\">\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-3 dark:border-gray-700\">\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-3 dark:border-gray-700\">\u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftbody>\u003C\u002Ftable>\u003C\u002Fdiv>\u003C\u002Fsection>\n\u003Ch2 id=\"section-16\">Gde se jednostavan primer zaustavlja\u003C\u002Fh2>\n\u003Cp>Nema svaki domen jedan neosporni autoritet. Istorijsko istraživanje može sadržati konfliktne primarne izvore. Naučne tvrdnje mogu se razvijati kako se pojavljuju nove studije. Pravno tumačenje može zavisiti od jurisdikcije, datuma i autoriteta suda. Ponašanje proizvoda može se razlikovati između dokumentacije i implementiranog koda.\u003C\u002Fp>\n\u003Cp>U tim slučajevima ispravna arhitektura nije da se izmisli jedan pobednik. Već da se očuvaju konkurentski izvori, njihova provenijencija, njihova klasa autoriteta, njihov primenljivi obim i nerešena kontradikcija. Pouzdan sistem Izvora istine mora biti sposoban da predstavi nesigurnost i neslaganje.\u003C\u002Fp>\n\u003Ch2 id=\"section-19\">Nadležnost je ograničena tvrdnjom, verzijom i vremenom\u003C\u002Fh2>\n\u003Cdiv class=\"overflow-x-auto\">\u003Ctable class=\"w-full border-collapse\">\u003Cthead>\u003Ctr>\u003Cth class=\"border border-gray-300 px-4 py-2 text-left font-semibold\">Pitanje\u003C\u002Fth>\u003Cth class=\"border border-gray-300 px-4 py-2 text-left font-semibold\">Mogući merodavni izvor\u003C\u002Fth>\u003Cth class=\"border border-gray-300 px-4 py-2 text-left font-semibold\">Zašto je opseg važan\u003C\u002Fth>\u003C\u002Ftr>\u003C\u002Fthead>\u003Ctbody>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Koje je trenutno stanje na korisničkom računu?\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Knjiga \u002F računovodstveni sistem evidencije\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Istorijski izvozi mogu biti tačni za ranije vreme, ali ne i za trenutno stanje.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Šta trenutno dozvoljava politika kompanije?\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Odobrena trenutna verzija politike\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Starija politika može ostati validan dokaz o prošlim pravilima, ali ne i o sadašnjim.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Koji je kod zaista raspoređen?\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Artefakt raspoređivanja \u002F commit \u002F zapis o izdanju\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Glavna grana se može razlikovati od produkcije.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Šta je ugovor navodio u trenutku potpisivanja?\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Izvršena verzija ugovora\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Nacrt ili kasniji šablon nije merodavan za potpisani sporazum.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Šta specificira tehnički protokol?\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Trenutna zvanična specifikacija za relevantnu verziju\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Objašnjenje na blogu može biti korisno, ali je sekundarni dokaz.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Šta se dogodilo u istorijskom događaju?\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Relevantni primarni dokazi plus eksplicitna kritika izvora\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Možda ne postoji jedinstveni autoritet; suprotstavljeni dokazi moraju ostati vidljivi.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Šta korisnik preferira?\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Trenutna eksplicitna korisnička postavka ili potvrđena preferencija\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Staro pamćenje razgovora može biti zastarelo ili zamenjeno.\u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftbody>\u003C\u002Ftable>\u003C\u002Fdiv>\n\u003Cp>Reč „istina“ stoga može biti obmanjujuća ako se ne navede njena granica. U arhitekturi, Izvor istine se obično bolje razume kao izvor ovlašćen da utvrdi određenu tvrdnju pod definisanim uslovima.\u003C\u002Fp>\n\u003Ch2 id=\"section-22\">Šta Izvor istine nije\u003C\u002Fh2>\n\u003Ch3 id=\"section-23\">Izvor istine naspram sistema evidencije\u003C\u002Fh3>\n\u003Cp>Sistem evidencije je obično merodavni operativni sistem za klasu zapisa: na primer, knjiga naplate, matični zapis ljudskih resursa ili baza podataka narudžbina. To je jedna uobičajena implementacija autoriteta Izvora istine.\u003C\u002Fp>\n\u003Cp>Ali Izvor istine je širi. Potpisani PDF ugovor, zvanični standard, artefakt raspoređivanja ili primarni arhivski dokument mogu biti merodavni, a da nisu transakcioni sistem evidencije.\u003C\u002Fp>\n\u003Ch3 id=\"section-26\">Izvor istine naspram provenijencije\u003C\u002Fh3>\n\u003Cp>Provenijencija odgovara na pitanja kao što su: Odakle potiču ovi podaci? Ko ili šta ih je proizveo? Koja transformacija je stvorila ovaj derivat? Koji prethodni entitet je korišćen? W3C PROV modeluje entitete, aktivnosti, agente i derivacije kako bi se poreklo i odgovornost mogli predstaviti.\u003C\u002Fp>\n\u003Cp>Provenijencija sama po sebi ne uspostavlja autoritet. Znati da vrednost potiče iz tabele koju je napisao određeni zaposleni pomaže u njenoj proceni, ali aplikacija i dalje zahteva pravilo koje kaže da li je ta tabela merodavna za tvrdnju.\u003C\u002Fp>\n\u003Ch3 id=\"section-29\">Izvor istine naspram dokaza\u003C\u002Fh3>\n\u003Cp>Dokaz podržava ili opovrgava tvrdnju. Izvor istine definiše koji izvor ima autoritet da reši ili snažno ograniči tu tvrdnju u trenutnom kontekstu aplikacije.\u003C\u002Fp>\n\u003Cp>Izvor može biti vredan dokaz, a da ne bude merodavan. Pet korisničkih imejlova može biti dokaz da korisnici ne vole radni tok, ali oni nisu sistem evidencije za trenutnu konfiguraciju proizvoda.\u003C\u002Fp>\n\u003Ch3 id=\"section-32\">Izvor istine naspram RAG-a\u003C\u002Fh3>\n\u003Cp>RAG je obrazac pretraživanja. Pronalazi informacije i dostavlja izabrani sadržaj modelu. RAG automatski ne zna koji izvor zaslužuje autoritet.\u003C\u002Fp>\n\u003Cp>RAG tok može da pronađe zastareo dokument, sekundarni rezime ili veoma sličan ali ne-merodavan izvor. Autoritet izvora mora biti kodiran kroz dizajn korpusa, metapodatke, filtere, politiku rangiranja, validaciju ili provere nakon pretraživanja.\u003C\u002Fp>\n\u003Ch3 id=\"section-35\">Izvor istine naspram vektorske baze podataka\u003C\u002Fh3>\n\u003Cp>Vektorska baza podataka čuva ili indeksira reprezentacije koje se koriste za semantičko pretraživanje. To je sloj pristupa, a ne automatski sloj istine.\u003C\u002Fp>\n\u003Cp>Isti autoritativni dokument može biti deljen u delove, ugrađen, kopiran i ponovo indeksiran mnogo puta. Vektorski zapis treba da zadrži referencu nazad na autoritativni izvor i verziju, umesto da postane novi autoritet bez mogućnosti praćenja.\u003C\u002Fp>\n\u003Ch3 id=\"section-38\">Izvor istine naspram memorije\u003C\u002Fh3>\n\u003Cp>Memorija agenta ili aplikacije čuva informacije koje mogu biti korisne kasnije. Memorija može sačuvati prethodnu odluku, preferenciju ili zapažanje, ali može postati zastarela.\u003C\u002Fp>\n\u003Cp>Za promenljivo ili posledično stanje, pouzdan agent bi normalno trebalo ponovo da pročita autoritativni trenutni izvor, umesto da pretpostavi da je zapamćeno stanje još uvek tačno.\u003C\u002Fp>\n\u003Ch3 id=\"section-41\">Izvor istine naspram konteksta\u003C\u002Fh3>\n\u003Cp>Kontekst je ono što model prima tokom trenutne inferencije. Autoritativne informacije mogu biti odsutne iz konteksta, dok neautoritativne informacije mogu biti prisutne.\u003C\u002Fp>\n\u003Cp>Izgradnja konteksta stoga zahteva politiku svesnu autoriteta: preuzeti ili pročitati izvor koji je ovlašćen da definiše tvrdnju, zatim sačuvati dovoljno metapodataka da model ili validator mogu da razumeju njen obim.\u003C\u002Fp>\n\u003Ch3 id=\"section-44\">Izvor istine naspram evaluacione osnovne istine\u003C\u002Fh3>\n\u003Cp>Evaluaciona osnovna istina je referentni odgovor, oznaka ili ishod prema kojem se sistem ocenjuje. Može se izvesti iz autoritativnih izvora, stručnog presuđivanja ili pripremljenih test podataka.\u003C\u002Fp>\n\u003Cp>Osnovna istina je stoga evaluacioni konstrukt. Izvor istine je konstrukt autoriteta aplikacije\u002Fdomena. Mogu se preklapati, ali nisu zamenljivi.\u003C\u002Fp>\n\u003Ch3 id=\"section-47\">Izvor istine naspram kvaliteta podataka\u003C\u002Fh3>\n\u003Cp>Autoritativni izvor i dalje može sadržati greške. Autoritet govori koji izvor zvanično upravlja činjenicom; kvalitet podataka pita da li je taj izvor tačan, potpun, pravovremen, konzistentan i prikladan za svrhu.\u003C\u002Fp>\n\u003Cp>Kada se zna da je autoritativni sistem pogrešan, arhitektura treba da zabeleži defekt, proces ispravke ili izuzetak, umesto da tiho zameni nezvaničnim izvorom i sakrije neslaganje.\u003C\u002Fp>\n\u003Ch2 id=\"section-50\">Praktičan arhitektonski model izvora istine\u003C\u002Fh2>\n\u003Caside class=\"editorjs-callout editorjs-callout--note my-6 rounded-xl border p-5 border-gray-300 bg-gray-50 dark:border-gray-700 dark:bg-gray-900\u002F40\" role=\"note\">\u003Cstrong class=\"block mb-2 text-gray-900 dark:text-gray-100\">Predloženi arhitektonski model\u003C\u002Fstrong>\u003Cdiv class=\"text-gray-700 dark:text-gray-200\">Model ispod je praktična sinteza za AI sisteme. Nije W3C, NIST ili ISO standard. Razdvaja autoritet, poreklo, preuzimanje i kontekst okrenut modelu, jer se te odgovornosti često pogrešno spajaju.\u003C\u002Fdiv>\u003C\u002Faside>\n\u003Csection class=\"editorjs-process my-6\">\u003Ch3 class=\"mb-3 text-lg font-semibold\">Put AI odgovora svestan autoriteta\u003C\u002Fh3>\u003Cdiv class=\"grid grid-cols-1 md:grid-cols-2 xl:grid-cols-3 gap-4\">\u003Cdiv class=\"editorjs-process__step min-w-0  rounded-xl border border-gray-200 dark:border-gray-700 p-4\">\u003Cdiv class=\"text-xs font-semibold text-gray-500 dark:text-gray-400\">1\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 font-semibold text-gray-900 dark:text-gray-100\">1. Definiši tip tvrdnje\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 text-sm text-gray-600 dark:text-gray-300\">Identifikuj šta korisnik zapravo pita: trenutno stanje, politiku, istorijsku činjenicu, tehničku specifikaciju, korisničku preferenciju, izračunavanje ili interpretaciju.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv class=\"editorjs-process__step min-w-0  rounded-xl border border-gray-200 dark:border-gray-700 p-4\">\u003Cdiv class=\"text-xs font-semibold text-gray-500 dark:text-gray-400\">2\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 font-semibold text-gray-900 dark:text-gray-100\">2. Razreši autoritet\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 text-sm text-gray-600 dark:text-gray-300\">Utvrdi koji izvor ili klasa autoriteta je ovlašćena da definiše taj tip tvrdnje za zahtevani obim, verziju i vreme.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv class=\"editorjs-process__step min-w-0  rounded-xl border border-gray-200 dark:border-gray-700 p-4\">\u003Cdiv class=\"text-xs font-semibold text-gray-500 dark:text-gray-400\">3\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 font-semibold text-gray-900 dark:text-gray-100\">3. Pribavi dokaze\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 text-sm text-gray-600 dark:text-gray-300\">Pročitaj ili preuzmi autoritativni izvor i sve potrebne prateće ili suprotstavljene dokaze.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv class=\"editorjs-process__step min-w-0  rounded-xl border border-gray-200 dark:border-gray-700 p-4\">\u003Cdiv class=\"text-xs font-semibold text-gray-500 dark:text-gray-400\">4\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 font-semibold text-gray-900 dark:text-gray-100\">4. Sačuvaj poreklo\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 text-sm text-gray-600 dark:text-gray-300\">Prenesi identitet izvora, verziju, vremensku oznaku, lokator, istoriju transformacija i metapodatke o odgovornosti.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv class=\"editorjs-process__step min-w-0  rounded-xl border border-gray-200 dark:border-gray-700 p-4\">\u003Cdiv class=\"text-xs font-semibold text-gray-500 dark:text-gray-400\">5\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 font-semibold text-gray-900 dark:text-gray-100\">5. Izgradi kontekst modela\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 text-sm text-gray-600 dark:text-gray-300\">Dostavi relevantne dokaze modelu bez odbacivanja metapodataka o autoritetu i primenljivosti.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv class=\"editorjs-process__step min-w-0  rounded-xl border border-gray-200 dark:border-gray-700 p-4\">\u003Cdiv class=\"text-xs font-semibold text-gray-500 dark:text-gray-400\">6\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 font-semibold text-gray-900 dark:text-gray-100\">6. Generiši ili izračunaj\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 text-sm text-gray-600 dark:text-gray-300\">Model može da sumira, uporedi, zaključuje ili transformiše dokaze, ali ne nasleđuje autoritet izvora samo obradom.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv class=\"editorjs-process__step min-w-0  rounded-xl border border-gray-200 dark:border-gray-700 p-4\">\u003Cdiv class=\"text-xs font-semibold text-gray-500 dark:text-gray-400\">7\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 font-semibold text-gray-900 dark:text-gray-100\">7. Validiraj tvrdnju\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 text-sm text-gray-600 dark:text-gray-300\">Proveri da je odgovor podržan od strane pravog izvora i da ostaje unutar svog obima i granice važenja.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv class=\"editorjs-process__step min-w-0  rounded-xl border border-gray-200 dark:border-gray-700 p-4\">\u003Cdiv class=\"text-xs font-semibold text-gray-500 dark:text-gray-400\">8\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 font-semibold text-gray-900 dark:text-gray-100\">8. Sačuvaj kontradikcije\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 text-sm text-gray-600 dark:text-gray-300\">Ako se autoritativni ili relevantni izvori ne slažu, izloži sukob umesto da izmišljaš lažnu sigurnost.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003C\u002Fsection>\n\u003Ch2 id=\"section-53\">Autoritet treba da bude eksplicitan, a ne izveden iz sličnosti\u003C\u002Fh2>\n\u003Cp>Jedan robustan obrazac implementacije je registar autoriteta ili ekvivalentan sloj politike koji mapira klase tvrdnji na klase autoritativnih izvora. Implementacija može biti kod, metapodaci, konfiguracija ili pravila domena; važno svojstvo je da je autoritet nameran.\u003C\u002Fp>\n\u003Cdiv class=\"overflow-x-auto\">\u003Ctable class=\"w-full border-collapse\">\u003Cthead>\u003Ctr>\u003Cth class=\"border border-gray-300 px-4 py-2 text-left font-semibold\">Klasa tvrdnje\u003C\u002Fth>\u003Cth class=\"border border-gray-300 px-4 py-2 text-left font-semibold\">Pravilo autoriteta\u003C\u002Fth>\u003Cth class=\"border border-gray-300 px-4 py-2 text-left font-semibold\">Rezervno ponašanje\u003C\u002Fth>\u003C\u002Ftr>\u003C\u002Fthead>\u003Ctbody>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Trenutno stanje računa\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Čitanje usluge aktivnog računa \u002F sistema evidencije\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Ako nije dostupno, prijaviti da trenutno stanje ne može biti verifikovano.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Dokumentacija proizvoda\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Trenutna odobrena verzija dokumentacije\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Starija verzija može biti prikazana samo uz upozorenje o verziji.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Implementirano ponašanje softvera\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Relevantno objavljeno izdanje \u002F izvorni artefakt\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Sama dokumentacija ne može dokazati objavljeno ponašanje.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Interna politika\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Odobreni repozitorijum politika i aktivna verzija\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Nacrti su prateći materijal, a ne trenutni autoritet.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Spoljni tehnički standard\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Publikacija zvaničnog standardizacionog tela za relevantnu verziju\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Sekundarna objašnjenja mogu pojasniti, ali ne mogu nadjačati specifikaciju.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Istraživačka tvrdnja\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Politika dokaza primjerena domenu\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Sačuvati suprotstavljene dokaze i pouzdanost umjesto forsiranja jednog izvora.\u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftbody>\u003C\u002Ftable>\u003C\u002Fdiv>\n\u003Ch2 id=\"section-56\">Pretraga treba da koristi autoritet kao ograničenje za rangiranje\u003C\u002Fh2>\n\u003Cp>Semantička relevantnost odgovara na pitanje „Koji kandidat izgleda povezano s ovim upitom?“ Autoritet odgovara na pitanje „Koji kandidat smije da utvrdi ovu činjenicu?“ Produkcioni sistem pretrage često zahtijeva oboje.\u003C\u002Fp>\n\u003Cp>Korisna sekvenca je prvo ograničiti prostor kandidata prema identitetu, zakupcu, klasi izvora, statusu, verziji ili datumu, a zatim rangirati relevantne dokaze unutar dozvoljenog prostora. Ako se relevantnost izračunava prije kritičnih autorizacijskih ili autoritetskih filtera, pipeline može vratiti uvjerljiv, ali nevažeći rezultat.\u003C\u002Fp>\n\u003Caside class=\"editorjs-callout editorjs-callout--warning my-6 rounded-xl border p-5 border-amber-300 bg-amber-50 dark:border-amber-900 dark:bg-amber-950\u002F20\" role=\"note\">\u003Cstrong class=\"block mb-2 text-gray-900 dark:text-gray-100\">Najveća sličnost nije najveći autoritet\u003C\u002Fstrong>\u003Cdiv class=\"text-gray-700 dark:text-gray-200\">Pretraživač može rangirati zastarjeli sažetak iznad trenutnog primarnog izvora. Jezički model može preferirati bolje napisano objašnjenje umjesto zvaničnog zapisa. Nijedno ponašanje ne mijenja koji izvor posjeduje činjenicu.\u003C\u002Fdiv>\u003C\u002Faside>\n\u003Ch2 id=\"section-60\">Svježina je dio autoriteta\u003C\u002Fh2>\n\u003Cp>Mnogi kvarovi izvora istine su zapravo vremenski kvarovi. Ispravan izvor je bio poznat, ali sistem je koristio stari snimak, zastarjeli embedding, keširani API odgovor ili zamijenjeni dokument.\u003C\u002Fp>\n\u003Cp>Pravilo autoriteta stoga treba uključivati semantiku poništavanja ili osvježavanja tamo gdje se činjenica može promijeniti. „CRM je autoritativan“ je nepotpuno kada aplikacija čita sedmicu star replicirani izvoz.\u003C\u002Fp>\n\u003Ch2 id=\"section-63\">Izvedene vrijednosti zahtijevaju trag natrag do autoritativnih ulaza\u003C\u002Fh2>\n\u003Cp>Neke važne činjenice nisu pohranjene direktno. One se izračunavaju iz autoritativnih ulaza: rezultat rizika, ukupan iznos računa, status podobnosti ili agregirana metrika.\u003C\u002Fp>\n\u003Cp>Za izvedene vrijednosti, arhitektura izvora istine treba da sačuva ulazne autoritete, verziju transformacije ili izračuna i vrijeme izvršenja. W3C PROV razlikovanje između entiteta, aktivnosti i derivacija je ovdje korisno jer modeluje kako je jedan entitet proizveden iz drugih.\u003C\u002Fp>\n\u003Caside class=\"editorjs-callout editorjs-callout--success my-6 rounded-xl border p-5 border-emerald-300 bg-emerald-50 dark:border-emerald-900 dark:bg-emerald-950\u002F20\" role=\"note\">\u003Cstrong class=\"block mb-2 text-gray-900 dark:text-gray-100\">Izvedena istina je uslovna\u003C\u002Fstrong>\u003Cdiv class=\"text-gray-700 dark:text-gray-200\">Izračunata vrijednost je samo onoliko aktuelna i autoritativna koliko su njeni ulazi, transformacija i uslovi važenja. Sačuvajte lozu potrebnu za ponovni izračun ili reviziju.\u003C\u002Fdiv>\u003C\u002Faside>\n\u003Ch2 id=\"section-67\">Šta se dešava kada se autoritativni izvori ne slažu?\u003C\u002Fh2>\n\u003Cp>Sukobi nisu rubni slučajevi u ozbiljnim sistemima znanja. Potpisan ugovor može se ne slagati s poljem u CRM-u. Produkciono ponašanje može se ne slagati s dokumentacijom. Dva primarna istorijska izvora mogu protivrječiti jedan drugom. Trenutna politika može biti u sukobu sa zastarjelom lokalnom kopijom.\u003C\u002Fp>\n\u003Cp>Sistem zahtijeva politiku rješavanja primjerenu domenu. Ponekad jedan autoritet jasno nadjačava drugi. Ponekad novija verzija zamjenjuje stariju. Ponekad stručnjak ili vlasnik poslovanja mora presuditi. A ponekad je ispravan rezultat jednostavno: dokazi su neriješeni.\u003C\u002Fp>\n\u003Cdiv class=\"overflow-x-auto\">\u003Ctable class=\"w-full border-collapse\">\u003Cthead>\u003Ctr>\u003Cth class=\"border border-gray-300 px-4 py-2 text-left font-semibold\">Tip sukoba\u003C\u002Fth>\u003Cth class=\"border border-gray-300 px-4 py-2 text-left font-semibold\">Tipično postupanje\u003C\u002Fth>\u003C\u002Ftr>\u003C\u002Fthead>\u003Ctbody>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Trenutna vs zamijenjena verzija\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Koristiti trenutnu verziju za sadašnje stanje; zadržati stariju verziju kao istorijski dokaz.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Sistem evidencije vs zastarjela replika\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Koristiti sistem evidencije; označiti problem svježine replikacije.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Ugovor vs CRM transkripcija\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Izvršeni ugovor upravlja ugovornim tekstom; CRM neslaganje postaje zadatak ispravke.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Dokumentacija vs objavljeno ponašanje\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Razlikovati namjeravano ponašanje od posmatranog\u002Fobjavljenog ponašanja; ne spajati ih tiho.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Dva vjerodostojna primarna izvora\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Sačuvati oba, procijeniti porijeklo i obim, i predstaviti neriješeno neslaganje ako ne postoji nadležni autoritet.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Korisničko pamćenje vs trenutna korisnička postavka\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Koristiti trenutnu eksplicitnu postavku; označiti pamćenje kao zamijenjeno gdje je prikladno.\u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftbody>\u003C\u002Ftable>\u003C\u002Fdiv>\n\u003Ch2 id=\"section-71\">Web pretraga je otkrivanje, ne automatski dokaz\u003C\u002Fh2>\n\u003Cp>Pretraživači su odlični sistemi za otkrivanje. Isječci pretrage, rangiranje rezultata i generisani sažeci nisu automatski primarni dokaz.\u003C\u002Fp>\n\u003Cp>Za tvrdnje koje zahtevaju autoritet, rezultat pretrage treba da vodi do originalne publikacije, zvaničnog zapisa, izvornog dokumenta, skupa podataka ili drugog odgovarajućeg artefakta. Stranica rezultata pomaže u lociranju izvora; ona ne nasleđuje autoritet izvora.\u003C\u002Fp>\n\u003Ch2 id=\"section-74\">Jezički model ne treba sam da odlučuje o autoritetu\u003C\u002Fh2>\n\u003Cp>Model može pomoći u klasifikaciji pitanja, izdvajanju tvrdnji ili upoređivanju dokaza, ali autoritet ne treba da zavisi samo od preferencije modela. Modeli optimizuju generisanje iz konteksta; oni ne poseduju garantovani domen-specifični registar koji određuje koja baza podataka, dokument ili organizacija poseduje svaku činjenicu.\u003C\u002Fp>\n\u003Cp>Zbog toga arhitektura aplikacije treba da kodira kritična pravila autoriteta deterministički gde je to praktično. Model može da rezonuje unutar granice, ali sama granica ne treba da se iznova stvara za svaki prompt.\u003C\u002Fp>\n\u003Ch2 id=\"section-77\">Autoritet mora da preživi trag izvršavanja\u003C\u002Fh2>\n\u003Cp>Ako je odgovor u produkciji dovoljno važan za reviziju, trag treba da omogući rekonstrukciju toga koji su izvori konsultovani, koja verzija je korišćena, koji odlomak ili zapis je podržao tvrdnju, koje transformacije su se dogodile i da li su bili dostupni suprotstavljeni dokazi.\u003C\u002Fp>\n\u003Cp>Ovo je u skladu sa širim principom provenijencije u W3C PROV i sa smernicama NIST AI RMF Playbook za dokumentovanje izvora, porekla, transformacija, zavisnosti, ograničenja i metapodataka.\u003C\u002Fp>\n\u003Ch2 id=\"section-80\">Dokaz originalne implementacije: Source of Truth Research Engine\u003C\u002Fh2>\n\u003Caside class=\"editorjs-callout editorjs-callout--note my-6 rounded-xl border p-5 border-gray-300 bg-gray-50 dark:border-gray-700 dark:bg-gray-900\u002F40\" role=\"note\">\u003Cstrong class=\"block mb-2 text-gray-900 dark:text-gray-100\">Dokaz originalne implementacije\u003C\u002Fstrong>\u003Cdiv class=\"text-gray-700 dark:text-gray-200\">Source of Truth Research Engine je moj sopstveni dokaz implementacije za arhitekturu koja stavlja dokaze na prvo mesto. On demonstrira jedan konkretan način da se razdvoje otkrivanje, pribavljanje izvora, provenijencija, tvrdnje, klase dokaza, kontradikcije i zaključci. To je obrazac implementacije, a ne univerzalni industrijski standard.\u003C\u002Fdiv>\u003C\u002Faside>\n\u003Cp>Engine je dizajniran oko sledivog pipeline-a, a ne direktnog AI sumiranja: istraživački zadatak → pretraga → originalni izvor ili digitalni artefakt → lokalni snimak → SHA-256 → ID izvora → tvrdnja → klasa dokaza → relacija ili kontradikcija → interpretacija → zaključak.\u003C\u002Fp>\n\u003Cp>Njegovo zajedničko jezgro dokaza čuva Sources, Artifacts, provenijenciju, Claims, Relations, Contradictions, Reference Model i revizorski trag. Različiti režimi istraživanja mogu da dele to jezgro dok primenjuju različite domenske metodologije.\u003C\u002Fp>\n\u003Cp>Arhitektura namerno razdvaja otkrivanje od dokaza. Isječci pretrage se ne tretiraju kao dokaz, nazivi datoteka se ne tretiraju kao sadržaj, AI sažeci se ne tretiraju kao primarni izvori, a semantička sličnost je samo signal za otkrivanje dok se rezultat ne poveže nazad sa konkretnim izvorom i lokatorom.\u003C\u002Fp>\n\u003Cp>Originalne datoteke se čuvaju, a lokalni bajtovi dobijaju SHA-256 identifikatore. Kontradikcije i odbačene hipoteze se ne brišu tiho. Novim dokazima je dozvoljeno da promene trenutni referentni model, dok prethodni put dokaza ostaje podložan reviziji.\u003C\u002Fp>\n\u003Cdiv class=\"overflow-x-auto\">\u003Ctable class=\"w-full border-collapse\">\u003Cthead>\u003Ctr>\u003Cth class=\"border border-gray-300 px-4 py-2 text-left font-semibold\">Implementirano pravilo\u003C\u002Fth>\u003Cth class=\"border border-gray-300 px-4 py-2 text-left font-semibold\">Zašto je važno za arhitekturu izvora istine\u003C\u002Fth>\u003C\u002Ftr>\u003C\u002Fthead>\u003Ctbody>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Pretraga ≠ dokaz\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Rangiranje otkrivanja ne može tiho da postane autoritet.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Naziv datoteke ≠ sadržaj\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Tragovi u metapodacima ne mogu da zamene čitanje stvarnog artefakta.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Lokalni snimak + SHA-256\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Dokaz može da se veže za tačne bajtove umesto za promenljivu udaljenu oznaku.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">ID izvora + tačan lokator\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Tvrdnje mogu da se prate nazad do konkretne lokacije dokaza.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Razdvajanje tvrdnje i dokaza\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Tvrdnja se ne meša sa materijalom koji je podržava.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Kontradikcije sačuvane\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Sistem može da predstavi nerešeno neslaganje umesto da prepiše istoriju.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Semantička sličnost je samo otkrivanje\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Relevantnost pretrage je eksplicitno odvojena od dokaznog autoriteta.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Novi dokazi mogu da ažuriraju model\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Stanje izvora istine je verzionisano i podložno reviziji, a ne tretira se kao nepromenljiva dogma.\u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftbody>\u003C\u002Ftable>\u003C\u002Fdiv>\n\u003Ch2 id=\"section-87\">Aaasaasa Document & Knowledge Engine: primena iste granice na poslovne dokumente\u003C\u002Fh2>\n\u003Cp>Koncept Aaasaasa Document & Knowledge Engine proširuje isti princip dizajna na poslovnu dokumentaciju: korisnici treba da mogu da pretražuju dokumente, postavljaju pitanja zasnovana na izvorima i pregledaju kolekcije prema eksplicitnim kriterijumima, uz očuvanje razlike između onoga što dokument navodi i onoga što sistem zaključuje.\u003C\u002Fp>\n\u003Cp>Važno arhitektonsko pravilo je da univerzalno jezgro za pretragu ne čini svaku kolekciju jednako autoritativnom. Ugovorni dokumenti, evidencija održavanja, finansijski dokumenti i istraživački materijal zahtevaju različita pravila autoriteta, validacije i pokrivenosti čak i kada dele infrastrukturu za unos i pretragu.\u003C\u002Fp>\n\u003Ch2 id=\"section-90\">Uobičajeni načini neuspeha izvora istine\u003C\u002Fh2>\n\u003Cdiv class=\"overflow-x-auto\">\u003Ctable class=\"w-full border-collapse\">\u003Cthead>\u003Ctr>\u003Cth class=\"border border-gray-300 px-4 py-2 text-left font-semibold\">Način neuspeha\u003C\u002Fth>\u003Cth class=\"border border-gray-300 px-4 py-2 text-left font-semibold\">Šta ide naopako\u003C\u002Fth>\u003C\u002Ftr>\u003C\u002Fthead>\u003Ctbody>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Model se tretira kao izvor istine\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Parametarsko znanje može biti zastarelo, nepotpuno, neproverljivo ili izvan autoritativnog domena aplikacije.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Najbolji rezultat pretrage automatski pobeđuje\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Sličnost se pogrešno smatra autoritetom.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Vektorska baza podataka postaje autoritativna\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Izvedeni zapisi indeksa gube identitet i verziju originalnog izvora.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Sve se kopira u jednu bazu znanja\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Kopije prikrivaju vlasništvo, svežinu i puteve ispravke.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Memorija se ponovo koristi kao trenutno stanje\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Stara zapažanja tiho nadjačavaju trenutni sistem evidencije.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Nema metapodataka o verziji\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Ispravan dokument se koristi za pogrešan vremenski period.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Nema lokatora izvora\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Citiranje postoji, ali prateći odlomak ili zapis ne može se proveriti.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Konflikti se prepisuju\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Sistem izgleda konzistentno uništavanjem dokaza o neslaganju.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Generisani rezimei zamenjuju originale\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Transformacija sa gubicima postaje prividni autoritet.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Autoritet je globalan umesto specifičan za tvrdnju\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Jednom izvoru se veruje izvan domena ili klase činjenica koje zaista poseduje.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Web isečak se tretira kao dokaz\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Metapodaci za otkrivanje zamenjuju originalnu publikaciju.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Autoritativni podaci su pogrešni, ali izuzeci su skriveni\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Operativni defekti postaju nevidljivi i ne mogu se transparentno ispraviti.\u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftbody>\u003C\u002Ftable>\u003C\u002Fdiv>\n\u003Ch2 id=\"section-92\">Praktični okvir za odlučivanje o izvoru istine\u003C\u002Fh2>\n\u003Csection class=\"editorjs-process my-6\">\u003Ch3 class=\"mb-3 text-lg font-semibold\">Kako odlučiti šta treba da definiše tvrdnju\u003C\u002Fh3>\u003Cdiv class=\"grid grid-cols-1 md:grid-cols-2 xl:grid-cols-3 gap-4\">\u003Cdiv class=\"editorjs-process__step min-w-0  rounded-xl border border-gray-200 dark:border-gray-700 p-4\">\u003Cdiv class=\"text-xs font-semibold text-gray-500 dark:text-gray-400\">1\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 font-semibold text-gray-900 dark:text-gray-100\">1. Precizno navedite tvrdnju\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 text-sm text-gray-600 dark:text-gray-300\">Razdvojite trenutno stanje, istorijsko stanje, politiku, interpretaciju, predviđanje i izvedeni proračun.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv class=\"editorjs-process__step min-w-0  rounded-xl border border-gray-200 dark:border-gray-700 p-4\">\u003Cdiv class=\"text-xs font-semibold text-gray-500 dark:text-gray-400\">2\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 font-semibold text-gray-900 dark:text-gray-100\">2. Identifikujte vlasnika autoriteta\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 text-sm text-gray-600 dark:text-gray-300\">Odredite sistem, dokument, instituciju, osobu ili klasu dokaza odgovornu za taj tip tvrdnje.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv class=\"editorjs-process__step min-w-0  rounded-xl border border-gray-200 dark:border-gray-700 p-4\">\u003Cdiv class=\"text-xs font-semibold text-gray-500 dark:text-gray-400\">3\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 font-semibold text-gray-900 dark:text-gray-100\">3. Definišite obim\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 text-sm text-gray-600 dark:text-gray-300\">Navedite zakupca, jurisdikciju, proizvod, okruženje, korisnika, skup dokumenata ili drugu granicu primenljivosti.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv class=\"editorjs-process__step min-w-0  rounded-xl border border-gray-200 dark:border-gray-700 p-4\">\u003Cdiv class=\"text-xs font-semibold text-gray-500 dark:text-gray-400\">4\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 font-semibold text-gray-900 dark:text-gray-100\">4. Definišite vreme i verziju\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 text-sm text-gray-600 dark:text-gray-300\">Odredite da li tvrdnja zahteva trenutno stanje, istorijski snimak ili određenu verziju standarda\u002Fizdanja.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv class=\"editorjs-process__step min-w-0  rounded-xl border border-gray-200 dark:border-gray-700 p-4\">\u003Cdiv class=\"text-xs font-semibold text-gray-500 dark:text-gray-400\">5\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 font-semibold text-gray-900 dark:text-gray-100\">5. Sačuvajte provenijenciju\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 text-sm text-gray-600 dark:text-gray-300\">Zabeležite identitet izvora, poreklo, lokator, transformacije i odgovorne agente ili procese.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv class=\"editorjs-process__step min-w-0  rounded-xl border border-gray-200 dark:border-gray-700 p-4\">\u003Cdiv class=\"text-xs font-semibold text-gray-500 dark:text-gray-400\">6\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 font-semibold text-gray-900 dark:text-gray-100\">6. Definišite put pretrage\u002Fpristupa\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 text-sm text-gray-600 dark:text-gray-300\">Uverite se da aplikacija zaista može da dobije autoritativnu informaciju pod ispravnim identitetom i dozvolama.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv class=\"editorjs-process__step min-w-0  rounded-xl border border-gray-200 dark:border-gray-700 p-4\">\u003Cdiv class=\"text-xs font-semibold text-gray-500 dark:text-gray-400\">7\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 font-semibold text-gray-900 dark:text-gray-100\">7. Definišite politiku konflikta\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 text-sm text-gray-600 dark:text-gray-300\">Odlučite o prvenstvu, zameni, presuđivanju ili eksplicitnom ponašanju u nerazrešenom stanju.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv class=\"editorjs-process__step min-w-0  rounded-xl border border-gray-200 dark:border-gray-700 p-4\">\u003Cdiv class=\"text-xs font-semibold text-gray-500 dark:text-gray-400\">8\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 font-semibold text-gray-900 dark:text-gray-100\">8. Definišite poništavanje\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 text-sm text-gray-600 dark:text-gray-300\">Navedite kada se keširani, indeksirani, zapamćeni ili izvedeni prikazi moraju osvežiti.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv class=\"editorjs-process__step min-w-0  rounded-xl border border-gray-200 dark:border-gray-700 p-4\">\u003Cdiv class=\"text-xs font-semibold text-gray-500 dark:text-gray-400\">9\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 font-semibold text-gray-900 dark:text-gray-100\">9. Validirajte put odgovora\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 text-sm text-gray-600 dark:text-gray-300\">Proverite da se važne generisane tvrdnje mogu pratiti do predviđenog autoriteta, a ne samo do verodostojnog izvora.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003C\u002Fsection>\n\u003Ch2 id=\"section-94\">Kontrolna lista arhitekture izvora istine\u003C\u002Fh2>\n\u003Cdiv class=\"overflow-x-auto\">\u003Ctable class=\"w-full border-collapse\">\u003Cthead>\u003Ctr>\u003Cth class=\"border border-gray-300 px-4 py-2 text-left font-semibold\">Pitanje\u003C\u002Fth>\u003Cth class=\"border border-gray-300 px-4 py-2 text-left font-semibold\">Očekivani odgovor\u003C\u002Fth>\u003C\u002Ftr>\u003C\u002Fthead>\u003Ctbody>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Koja tačna činjenica ili stanje se utvrđuje?\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Tvrdnja dovoljno precizna da joj se dodeli autoritet.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Ko ili šta poseduje tu činjenicu?\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Imenovani autoritativni sistem, klasa izvora ili pravilo presuđivanja.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Da li je autoritet aktuelan za ovaj obim?\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Granica zakupca, jurisdikcije, okruženja, korisnika ili domena je eksplicitna.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Da li su verzija\u002Fvreme ispravni?\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Trenutna, istorijska ili verzijski specifična primenljivost je poznata.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Može li se izvor proveriti?\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Postoji stabilni identifikator, lokator ili referenca zapisa.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Da li je provenijencija sačuvana?\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Metapodaci o poreklu, transformaciji i odgovornosti preživljavaju unos i pretragu.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Može li pretraga vratiti neautoritativni materijal?\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Ako da, filteri ili validacija razlikuju relevantnost od autoriteta.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Može li se izvor promeniti?\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Postoje pravila za osvežavanje, poništavanje ili zamenu.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Mogu li se izvori ne slagati?\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Ponašanje pri konfliktu i presuđivanju je eksplicitno.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Može li memorija postati zastarela?\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Nestabilno stanje se ponovo čita iz trenutnog autoriteta pre posledične upotrebe.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Može li se izvedeni odgovor reprodukovati?\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Ulazi, verzija transformacije i uslovi izvršavanja su sledljivi.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Može li revizor rekonstruisati odgovor?\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Dokazi o izvršavanju čuvaju put izvora za važne tvrdnje.\u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftbody>\u003C\u002Ftable>\u003C\u002Fdiv>\n\u003Ch2 id=\"section-96\">Uobičajene zablude\u003C\u002Fh2>\n\u003Cdiv class=\"overflow-x-auto\">\u003Ctable class=\"w-full border-collapse\">\u003Cthead>\u003Ctr>\u003Cth class=\"border border-gray-300 px-4 py-2 text-left font-semibold\">Zabluda\u003C\u002Fth>\u003Cth class=\"border border-gray-300 px-4 py-2 text-left font-semibold\">Ispravka\u003C\u002Fth>\u003C\u002Ftr>\u003C\u002Fthead>\u003Ctbody>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">„Izvor istine znači jedna baza podataka.“\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Jedna baza podataka može biti autoritativna za jedan domen; složeni sistemi obično imaju više autoriteta specifičnih za činjenice.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">„Najnoviji dokument je automatski autoritativan.“\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Novina pomaže samo kada je noviji artefakt odobren i zaista zamenjuje stariji.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">„RAG rešava istinu.“\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">RAG rešava pretragu. Autoritet, provenijencija, kvalitet dokaza i validnost ostaju odvojeni problemi.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">„Citiranje dokazuje odgovor.“\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Citirani izvor mora zaista podržavati tvrdnju, imati pravi autoritet i primenjivati se na trenutni obim.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">„Provenijencija nam govori šta je istinito.“\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Provenijencija nam govori o poreklu i izvođenju; autoritet i ispravnost i dalje zahtevaju pravila domena i evaluaciju.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">„Sistem evidencije je uvek ispravan.“\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">On je autoritativan za operativni zapis, ali defekti kvaliteta podataka i dalje mogu postojati i zahtevaju vidljivu ispravku.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">„Ako se nekoliko izvora slaže, tvrdnja je autoritativna.“\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Slaganje povećava dokaze, ali ne uspostavlja nužno vlasništvo ili primenljivost.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">„AI memorija može zameniti ponovljena čitanja.“\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Samo za informacije čiji je rizik zastarevanja prihvatljiv; nestabilno ili posledično stanje treba osvežiti iz autoriteta.\u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftbody>\u003C\u002Ftable>\u003C\u002Fdiv>\n\u003Ch2 id=\"section-98\">Granični slučajevi\u003C\u002Fh2>\n\u003Cp>Neka pitanja su interpretativna, a ne činjenična. „Koja arhitektura je najbolja?“ nema jedinstveni izvor istine. Sistem može da dohvati autoritativna ograničenja i dokaze, ali konačni sud je zaključak koji treba da izloži pretpostavke i kompromise.\u003C\u002Fp>\n\u003Cp>Neki domeni koriste distribuiran autoritet. Naučni zaključak može zavisiti od više studija, skupova podataka i replikacija. Istorijski zaključak može zavisiti od suprotstavljenih primarnih i sekundarnih dokaza. Arhitektura treba da predstavi strukturu dokaza, a ne da izmišlja centralnu bazu podataka koja navodno poseduje istinu.\u003C\u002Fp>\n\u003Cp>Korisnik takođe može biti autoritet za subjektivne lične informacije: preferencije, ciljeve, izabrana podešavanja ili eksplicitna uputstva. Čak i tada, noviji eksplicitni unos može zameniti stariju memoriju.\u003C\u002Fp>\n\u003Cp>Spoljni događaji mogu poništiti prethodno autoritativne podatke. Cenovni feed, sistem zaliha ili bezbednosna politika mogli su biti ispravni kada su zabeleženi, ali više nisu važeći. Provenijencija snimka čuva ono što je tada bilo istinito; ona ne čini snimak zauvek aktuelnim.\u003C\u002Fp>\n\u003Ch2 id=\"section-103\">Ograničenja\u003C\u002Fh2>\n\u003Cp>Arhitektura izvora istine ne može garantovati da je autoritativni izvor činjenično ispravan. Ona obezbeđuje odgovornost, provenijenciju i determinističke granice vlasništva; procesi za kvalitet podataka i verifikaciju domena ostaju neophodni.\u003C\u002Fp>\n\u003Cp>Autoritet takođe može biti sporan. Različite institucije mogu legitimno polagati pravo na autoritet u različitim jurisdikcijama ili metodologijama. U takvim situacijama sistem treba da izloži model autoriteta i neslaganje, a ne da ih sakrije iza univerzalnog „rezultata istinitosti“.\u003C\u002Fp>\n\u003Cp>Konačno, pravila autoriteta zahtevaju održavanje. Sistemi, vlasnici, politike, verzije i propisi se menjaju. Zastareli registar autoriteta može biti jednako opasan kao i nikakav registar.\u003C\u002Fp>\n\u003Ch2 id=\"section-107\">Šta bi promenilo ovaj odgovor?\u003C\u002Fh2>\n\u003Cp>Specifično mapiranje autoriteta menja se u zavisnosti od domena. Bankarstvo, zdravstvo, isporuka softvera, naučno istraživanje i istorijska analiza imaju različite sisteme evidencije, dokazna pravila i regulatorne obaveze.\u003C\u002Fp>\n\u003Cp>Implementacija se takođe menja sa arhitekturom. Mala aplikacija može da kodira autoritet direktno u pozivima servisa. Veća platforma može zahtevati registre, metapodatke izvora, mehanizme politika, sisteme porekla ili ugovore o podacima. Osnovni princip ostaje isti: ne dozvolite da redosled pretrage ili preferencija modela tiho odlučuju šta se smatra autoritativnim.\u003C\u002Fp>\n\u003Ch2 id=\"section-110\">Povezano kanonsko znanje\u003C\u002Fh2>\n\u003Cp>Arhitektura izvora istine je preduslov za kasnije koncepte pretrage i upravljanja jer kvalitet pretrage sam po sebi ne može da odredi da li je dokazima dozvoljeno da definišu odgovor.\u003C\u002Fp>\n\u003Caside class=\"editorjs-referral my-6\">\u003Ca href=\"https:\u002F\u002Fstajic.de\u002Fsr\u002Fblog\u002Fwhat-is-rag-the-simplest-explanation-of-how-it-works\" class=\"flex flex-col sm:flex-row gap-4 rounded-xl border border-gray-200 dark:border-gray-700 p-4 transition hover:border-primary-500\">\u003Cdiv class=\"min-w-0 flex-1\">\u003Cstrong class=\"block text-lg text-gray-900 dark:text-gray-100\">Šta je RAG? Najjednostavnije objašnjenje kako funkcioniše\u003C\u002Fstrong>\u003Cp class=\"mt-2 text-sm text-gray-600 dark:text-gray-300\">RAG pronalazi eksterno znanje za model. Ovaj temelj objašnjava zašto su pretraga i autoritativna istina odvojene odgovornosti.\u003C\u002Fp>\u003Cspan class=\"mt-3 inline-flex text-sm font-medium text-primary-600 dark:text-primary-400\">Pročitajte osnove RAG-a →\u003C\u002Fspan>\u003C\u002Fdiv>\u003C\u002Fa>\u003C\u002Faside>\n\u003Cp>Memorija je još jedan srodan koncept. Pouzdan agent razdvaja zapamćene informacije od trenutnog autoritativnog stanja aplikacije.\u003C\u002Fp>\n\u003Caside class=\"editorjs-referral my-6\">\u003Ca href=\"https:\u002F\u002Fstajic.de\u002Fsr\u002Fblog\u002Fai-agent-memory-is-not-rag-how-to-separate-memory-retrieval-state-and-context\" class=\"flex flex-col sm:flex-row gap-4 rounded-xl border border-gray-200 dark:border-gray-700 p-4 transition hover:border-primary-500\">\u003Cdiv class=\"min-w-0 flex-1\">\u003Cstrong class=\"block text-lg text-gray-900 dark:text-gray-100\">Memorija AI agenta nije RAG: Kako razdvojiti memoriju, pretragu, stanje i kontekst\u003C\u002Fstrong>\u003Cp class=\"mt-2 text-sm text-gray-600 dark:text-gray-300\">Arhitektura sa četiri sloja koja razdvaja trajnu memoriju, autoritativno stanje, pretragu i kontekst koji se stvarno dostavlja modelu.\u003C\u002Fp>\u003Cspan class=\"mt-3 inline-flex text-sm font-medium text-primary-600 dark:text-primary-400\">Pročitajte članak o arhitekturi memorije →\u003C\u002Fspan>\u003C\u002Fdiv>\u003C\u002Fa>\u003C\u002Faside>\n\u003Cp>Autoritet se takođe direktno povezuje sa validnošću odgovora. Čak i autoritativni izvor podržava samo tvrdnje unutar svoje verzije, datuma, obima i granice dokaza.\u003C\u002Fp>\n\u003Caside class=\"editorjs-referral my-6\">\u003Ca href=\"https:\u002F\u002Fstajic.de\u002Fsr\u002Fblog\u002Fthe-answer-validity-boundary-the-missing-layer-between-relevance-and-reliable-ai-answers\" class=\"flex flex-col sm:flex-row gap-4 rounded-xl border border-gray-200 dark:border-gray-700 p-4 transition hover:border-primary-500\">\u003Cdiv class=\"min-w-0 flex-1\">\u003Cstrong class=\"block text-lg text-gray-900 dark:text-gray-100\">Granica validnosti odgovora: Sloj koji nedostaje između relevantnosti i pouzdanih AI odgovora\u003C\u002Fstrong>\u003Cp class=\"mt-2 text-sm text-gray-600 dark:text-gray-300\">Okvir za eksplicitno određivanje gde AI odgovor ostaje podržan dokazima i koje promenjene okolnosti poništavaju tu podršku.\u003C\u002Fp>\u003Cspan class=\"mt-3 inline-flex text-sm font-medium text-primary-600 dark:text-primary-400\">Pročitajte Granicu validnosti odgovora →\u003C\u002Fspan>\u003C\u002Fdiv>\u003C\u002Fa>\u003C\u002Faside>\n\u003Ch2 id=\"section-117\">Često postavljana pitanja\u003C\u002Fh2>\n\u003Csection class=\"editorjs-faq my-6 rounded-xl border border-gray-200 p-5 dark:border-gray-700\">\u003Ch3 class=\"mb-3 text-lg font-semibold\">Izvor istine u AI sistemima\u003C\u002Fh3>\u003Cdiv id=\"faq1\" class=\"border-t border-gray-200 py-4 first:border-t-0 dark:border-gray-700\">\u003Ch4 class=\"font-semibold text-gray-900 dark:text-gray-100\">Šta je izvor istine u AI sistemu?\u003C\u002Fh4>\u003Cdiv class=\"mt-2 text-gray-600 dark:text-gray-300\">To je autoritativni izvor ili pravilo autoriteta koje određuje koji izvor sme da utvrdi određenu činjenicu, stanje ili pravilo za definisani obim, verziju i vreme.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv id=\"faq2\" class=\"border-t border-gray-200 py-4 first:border-t-0 dark:border-gray-700\">\u003Ch4 class=\"font-semibold text-gray-900 dark:text-gray-100\">Da li je jezički model izvor istine?\u003C\u002Fh4>\u003Cdiv class=\"mt-2 text-gray-600 dark:text-gray-300\">Obično ne. Jezički model može da generiše, sumira i zaključuje, ali njegovo parametarsko znanje nije automatski autoritativno za trenutno stanje aplikacije, politiku kompanije, određenu verziju dokumenta ili činjenicu iz regulisane oblasti.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv id=\"faq3\" class=\"border-t border-gray-200 py-4 first:border-t-0 dark:border-gray-700\">\u003Ch4 class=\"font-semibold text-gray-900 dark:text-gray-100\">Da li je vektorska baza podataka izvor istine za RAG?\u003C\u002Fh4>\u003Cdiv class=\"mt-2 text-gray-600 dark:text-gray-300\">Ne automatski. Vektorska baza podataka je obično indeks ili skladište za pretragu. Trebalo bi da čuva reference na autoritativni originalni izvor i verziju, a ne da ih tiho zamenjuje.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv id=\"faq4\" class=\"border-t border-gray-200 py-4 first:border-t-0 dark:border-gray-700\">\u003Ch4 class=\"font-semibold text-gray-900 dark:text-gray-100\">Koja je razlika između porekla i izvora istine?\u003C\u002Fh4>\u003Cdiv class=\"mt-2 text-gray-600 dark:text-gray-300\">Poreklo opisuje odakle podaci potiču, kako su proizvedeni ili transformisani i ko ili šta je bio uključen. Pravila izvora istine određuju da li taj izvor ima autoritet za konkretnu tvrdnju.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv id=\"faq5\" class=\"border-t border-gray-200 py-4 first:border-t-0 dark:border-gray-700\">\u003Ch4 class=\"font-semibold text-gray-900 dark:text-gray-100\">Može li AI sistem imati više izvora istine?\u003C\u002Fh4>\u003Cdiv class=\"mt-2 text-gray-600 dark:text-gray-300\">Da. U složenim sistemima to je normalno jer različite činjenice pripadaju različitim autoritativnim sistemima ili klasama izvora.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv id=\"faq6\" class=\"border-t border-gray-200 py-4 first:border-t-0 dark:border-gray-700\">\u003Ch4 class=\"font-semibold text-gray-900 dark:text-gray-100\">Šta se dešava kada se dva autoritativna izvora ne slažu?\u003C\u002Fh4>\u003Cdiv class=\"mt-2 text-gray-600 dark:text-gray-300\">Sistemu je potrebno pravilo za rešavanje sukoba specifično za domen: prioritet, zamena verzije, stručno presuđivanje ili eksplicitno nerešeno stanje. Ne bi trebalo tiho da bira izvor koji model preferira.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv id=\"faq7\" class=\"border-t border-gray-200 py-4 first:border-t-0 dark:border-gray-700\">\u003Ch4 class=\"font-semibold text-gray-900 dark:text-gray-100\">Da li RAG garantuje da AI odgovor koristi izvor istine?\u003C\u002Fh4>\u003Cdiv class=\"mt-2 text-gray-600 dark:text-gray-300\">Ne. RAG pronalazi kandidate. Potrebni su metapodaci svesni autoriteta, filteri, politika izvora i validacija da bi se osiguralo da važne tvrdnje koriste ispravan izvor.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv id=\"faq8\" class=\"border-t border-gray-200 py-4 first:border-t-0 dark:border-gray-700\">\u003Ch4 class=\"font-semibold text-gray-900 dark:text-gray-100\">Može li izvor istine biti pogrešan?\u003C\u002Fh4>\u003Cdiv class=\"mt-2 text-gray-600 dark:text-gray-300\">Da. Autoritet i ispravnost su različita svojstva. Autoritativni sistem može sadržati defekt u kvalitetu podataka, koji treba transparentno ispraviti, a ne sakriti zamenom nezvaničnim izvorom.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003C\u002Fsection>\n\u003Ch2 id=\"section-119\">Pojmovnik\u003C\u002Fh2>\n\u003Csection class=\"editorjs-glossary my-6 rounded-xl border border-gray-200 dark:border-gray-700 p-5\">\u003Ch3 class=\"mb-3 text-lg font-semibold\">Ključni pojmovi izvora istine\u003C\u002Fh3>\u003Cdl>\u003Cdiv id=\"source-of-truth\" class=\"border-t border-gray-200 dark:border-gray-700 py-3 first:border-t-0\">\u003Cdt class=\"font-semibold text-gray-900 dark:text-gray-100\">Izvor istine\u003C\u002Fdt>\u003Cdd class=\"mt-1 text-gray-600 dark:text-gray-300\">Autoritativni izvor ili pravilo koje je dozvoljeno da utvrdi određenu činjenicu, stanje ili pravilo za definisani obim, verziju i vreme.\u003C\u002Fdd>\u003C\u002Fdiv>\u003Cdiv id=\"system-of-record\" class=\"border-t border-gray-200 dark:border-gray-700 py-3 first:border-t-0\">\u003Cdt class=\"font-semibold text-gray-900 dark:text-gray-100\">Sistem evidencije\u003C\u002Fdt>\u003Cdd class=\"mt-1 text-gray-600 dark:text-gray-300\">Autoritativni operativni sistem odgovoran za definisanu klasu zapisa ili trenutno poslovno stanje.\u003C\u002Fdd>\u003C\u002Fdiv>\u003Cdiv id=\"provenance\" class=\"border-t border-gray-200 dark:border-gray-700 py-3 first:border-t-0\">\u003Cdt class=\"font-semibold text-gray-900 dark:text-gray-100\">Poreklo\u003C\u002Fdt>\u003Cdd class=\"mt-1 text-gray-600 dark:text-gray-300\">Informacije koje opisuju izvor, izvođenje, transformacije, odgovorne agente i istoriju podataka ili drugog entiteta.\u003C\u002Fdd>\u003C\u002Fdiv>\u003Cdiv id=\"evidence\" class=\"border-t border-gray-200 dark:border-gray-700 py-3 first:border-t-0\">\u003Cdt class=\"font-semibold text-gray-900 dark:text-gray-100\">Dokaz\u003C\u002Fdt>\u003Cdd class=\"mt-1 text-gray-600 dark:text-gray-300\">Informacija ili artefakt koji podržava, opovrgava ili ograničava tvrdnju.\u003C\u002Fdd>\u003C\u002Fdiv>\u003Cdiv id=\"authority\" class=\"border-t border-gray-200 dark:border-gray-700 py-3 first:border-t-0\">\u003Cdt class=\"font-semibold text-gray-900 dark:text-gray-100\">Autoritet\u003C\u002Fdt>\u003Cdd class=\"mt-1 text-gray-600 dark:text-gray-300\">Pravilo aplikacije ili domena koje određuje koji izvor ima pravo da definiše konkretnu tvrdnju.\u003C\u002Fdd>\u003C\u002Fdiv>\u003Cdiv id=\"freshness\" class=\"border-t border-gray-200 dark:border-gray-700 py-3 first:border-t-0\">\u003Cdt class=\"font-semibold text-gray-900 dark:text-gray-100\">Svežina\u003C\u002Fdt>\u003Cdd class=\"mt-1 text-gray-600 dark:text-gray-300\">Da li reprezentacija ostaje dovoljno aktuelna za tvrdnju ili operaciju u kojoj se koristi.\u003C\u002Fdd>\u003C\u002Fdiv>\u003Cdiv id=\"supersession\" class=\"border-t border-gray-200 dark:border-gray-700 py-3 first:border-t-0\">\u003Cdt class=\"font-semibold text-gray-900 dark:text-gray-100\">Zamena verzije\u003C\u002Fdt>\u003Cdd class=\"mt-1 text-gray-600 dark:text-gray-300\">Eksplicitna zamena starije autoritativne verzije novijom uz očuvanje istorijske sledljivosti.\u003C\u002Fdd>\u003C\u002Fdiv>\u003Cdiv id=\"ground-truth\" class=\"border-t border-gray-200 dark:border-gray-700 py-3 first:border-t-0\">\u003Cdt class=\"font-semibold text-gray-900 dark:text-gray-100\">Osnovna istina\u003C\u002Fdt>\u003Cdd class=\"mt-1 text-gray-600 dark:text-gray-300\">Referentni odgovor, oznaka ili ishod koji se koristi za evaluaciju sistema; to je konstrukt za evaluaciju, a ne automatski izvor istine aplikacije.\u003C\u002Fdd>\u003C\u002Fdiv>\u003Cdiv id=\"lineage\" class=\"border-t border-gray-200 dark:border-gray-700 py-3 first:border-t-0\">\u003Cdt class=\"font-semibold text-gray-900 dark:text-gray-100\">Sledljivost\u003C\u002Fdt>\u003Cdd class=\"mt-1 text-gray-600 dark:text-gray-300\">Trag kako podaci ili izvedene vrednosti teku i transformišu se kroz izvore i korake obrade.\u003C\u002Fdd>\u003C\u002Fdiv>\u003Cdiv id=\"validity-boundary\" class=\"border-t border-gray-200 dark:border-gray-700 py-3 first:border-t-0\">\u003Cdt class=\"font-semibold text-gray-900 dark:text-gray-100\">Granica validnosti\u003C\u002Fdt>\u003Cdd class=\"mt-1 text-gray-600 dark:text-gray-300\">Uslovi obima, vremena, verzije, dokaza i pretpostavki unutar kojih tvrdnja ostaje podržana.\u003C\u002Fdd>\u003C\u002Fdiv>\u003C\u002Fdl>\u003C\u002Fsection>\n\u003Ch2 id=\"section-121\">Zaključak\u003C\u002Fh2>\n\u003Cp>Pouzdan AI ne dolazi od davanja više informacija modelu. Dolazi od znanja koje informacije smeju da definišu tvrdnju, čuvanja odakle te informacije potiču, pronalaženja ispravne verzije i držanja konačnog odgovora unutar obima izvora.\u003C\u002Fp>\n\u003Cp>Zato izvor istine, poreklo, pretraga, memorija i kontekst moraju ostati odvojeni koncepti. Izvor istine definiše autoritet. Poreklo objašnjava nastanak. Pretraga pronalazi kandidate. Memorija čuva izabrane prošle informacije. Kontekst je ono što model vidi. Generisanje pretvara te ulaze u izlaz.\u003C\u002Fp>\n\u003Cp>Kada ti slojevi ostanu eksplicitni, AI sistem može da uradi više od zvučanja verodostojno: važne tvrdnje mogu se pratiti nazad do izvora koji je zaista imao pravo da ih utvrdi.\u003C\u002Fp>\n\u003Ch2 id=\"section-125\">Primarni izvori i dokazi o implementaciji\u003C\u002Fh2>\n\u003Cp>Spoljni izvori ispod podržavaju tvrdnje o poreklu i upravljanju rizikom AI. Odsek o istraživačkom motoru izvora istine je originalni dokaz o implementaciji i eksplicitno je predstavljen kao jedan obrazac implementacije, a ne kao univerzalni standard.\u003C\u002Fp>\n\u003Ca href=\"https:\u002F\u002Fwww.w3.org\u002FTR\u002Fprov-dm\u002F\" target=\"_blank\" rel=\"noopener noreferrer\" class=\"editorjs-link-tool block border border-gray-200 dark:border-gray-700 rounded-lg p-4 transition text-gray-900 dark:text-gray-100 hover:border-primary-500 hover:bg-primary-50 dark:hover:bg-gray-900 hover:text-gray-900 dark:hover:text-gray-100\">\u003Cstrong class=\"block font-semibold\">W3C PROV-DM — PROV model podataka\u003C\u002Fstrong>\u003Cp class=\"text-sm text-gray-600 dark:text-gray-400\">W3C preporuka koja definiše model provenijencije nezavisan od domena oko entiteta, aktivnosti, agenata, izvođenja i odgovornosti.\u003C\u002Fp>\u003C\u002Fa>\n\u003Ca href=\"https:\u002F\u002Fwww.w3.org\u002Fgroups\u002Fwg\u002Fprov\u002Fpublications\u002F\" target=\"_blank\" rel=\"noopener noreferrer\" class=\"editorjs-link-tool block border border-gray-200 dark:border-gray-700 rounded-lg p-4 transition text-gray-900 dark:text-gray-100 hover:border-primary-500 hover:bg-primary-50 dark:hover:bg-gray-900 hover:text-gray-900 dark:hover:text-gray-100\">\u003Cstrong class=\"block font-semibold\">W3C radna grupa za provenijenciju — publikacije\u003C\u002Fstrong>\u003Cp class=\"text-sm text-gray-600 dark:text-gray-400\">Zvanični indeks W3C PROV preporuka i povezanih specifikacija za razmenu provenijencije i ograničenja.\u003C\u002Fp>\u003C\u002Fa>\n\u003Ca href=\"https:\u002F\u002Fwww.nist.gov\u002Fitl\u002Fai-risk-management-framework\" target=\"_blank\" rel=\"noopener noreferrer\" class=\"editorjs-link-tool block border border-gray-200 dark:border-gray-700 rounded-lg p-4 transition text-gray-900 dark:text-gray-100 hover:border-primary-500 hover:bg-primary-50 dark:hover:bg-gray-900 hover:text-gray-900 dark:hover:text-gray-100\">\u003Cstrong class=\"block font-semibold\">NIST okvir za upravljanje rizicima AI sistema\u003C\u002Fstrong>\u003Cp class=\"text-sm text-gray-600 dark:text-gray-400\">NIST dobrovoljni okvir za uključivanje razmatranja pouzdanosti i upravljanja rizicima tokom životnog ciklusa AI sistema; AI RMF 1.0 je trenutno u fazi revizije.\u003C\u002Fp>\u003C\u002Fa>\n\u003Ca href=\"https:\u002F\u002Fairc.nist.gov\u002Fairmf-resources\u002Fplaybook\u002F\" target=\"_blank\" rel=\"noopener noreferrer\" class=\"editorjs-link-tool block border border-gray-200 dark:border-gray-700 rounded-lg p-4 transition text-gray-900 dark:text-gray-100 hover:border-primary-500 hover:bg-primary-50 dark:hover:bg-gray-900 hover:text-gray-900 dark:hover:text-gray-100\">\u003Cstrong class=\"block font-semibold\">NIST AI RMF priručnik\u003C\u002Fstrong>\u003Cp class=\"text-sm text-gray-600 dark:text-gray-400\">Operativne smernice usklađene sa AI RMF, uključujući prakse dokumentovanja za provenijenciju podataka, izvore, poreklo, transformacije, zavisnosti, ograničenja i metapodatke.\u003C\u002Fp>\u003C\u002Fa>\n\u003Ca href=\"https:\u002F\u002Fairc.nist.gov\u002Fairmf-resources\u002Fplaybook\u002Fmeasure\u002F\" target=\"_blank\" rel=\"noopener noreferrer\" class=\"editorjs-link-tool block border border-gray-200 dark:border-gray-700 rounded-lg p-4 transition text-gray-900 dark:text-gray-100 hover:border-primary-500 hover:bg-primary-50 dark:hover:bg-gray-900 hover:text-gray-900 dark:hover:text-gray-100\">\u003Cstrong class=\"block font-semibold\">NIST AI RMF priručnik — merenje\u003C\u002Fstrong>\u003Cp class=\"text-sm text-gray-600 dark:text-gray-400\">Smernice o dokumentovanju merenja, provenijencije podataka i kontekstualne interpretacije izlaza AI sistema.\u003C\u002Fp>\u003C\u002Fa>\n\u003Ca href=\"https:\u002F\u002Fwww.nist.gov\u002Fpublications\u002Fartificial-intelligence-risk-management-framework-generative-artificial-intelligence\" target=\"_blank\" rel=\"noopener noreferrer\" class=\"editorjs-link-tool block border border-gray-200 dark:border-gray-700 rounded-lg p-4 transition text-gray-900 dark:text-gray-100 hover:border-primary-500 hover:bg-primary-50 dark:hover:bg-gray-900 hover:text-gray-900 dark:hover:text-gray-100\">\u003Cstrong class=\"block font-semibold\">NIST AI 600-1 — profil generativne veštačke inteligencije\u003C\u002Fstrong>\u003Cp class=\"text-sm text-gray-600 dark:text-gray-400\">NIST profil generativne veštačke inteligencije, uključujući razmatranja provenijencije i integriteta informacija za sisteme generativne veštačke inteligencije.\u003C\u002Fp>\u003C\u002Fa>",{"time":212,"blocks":213,"version":1296},1791479247666,[214,220,228,235,241,249,254,259,264,269,276,281,286,291,296,331,336,341,346,351,388,393,398,403,408,413,418,423,428,433,438,443,448,453,458,463,468,473,478,483,488,493,498,503,508,513,518,523,528,533,538,544,576,581,586,619,624,629,634,641,646,651,656,661,666,671,677,682,687,692,718,723,728,733,738,743,748,753,758,763,768,774,779,784,789,794,826,831,836,841,846,890,895,928,933,976,981,1013,1018,1023,1028,1033,1038,1043,1048,1053,1058,1063,1068,1073,1078,1083,1092,1097,1105,1110,1118,1123,1161,1166,1211,1216,1221,1226,1231,1236,1241,1251,1260,1269,1278,1287],{"id":215,"data":216,"type":218,"tunes":219},"intro",{"text":217},"Izvor istine u AI sistemu je autoritativni izvor koji je ovlašćen da definiše da li određena činjenica, stanje ili pravilo treba da se tretiraju kao istiniti za određeni obim, verziju i vreme. To automatski nije jezički model, vektorska baza podataka, najbolje rangirani preuzeti dokument, memorija agenta ili najnovija poruka u kontekstu. Pouzdana AI arhitektura mora da očuva koji izvor ima autoritet za koju tvrdnju, a zatim da održi provenijenciju, preuzimanje i validaciju povezane sa tim autoritetom.","paragraph",{},{"id":221,"data":222,"type":226,"tunes":227},"direct",{"body":223,"title":224,"variant":225},"\u003Cstrong>Izvor istine je pravilo autoriteta, a ne AI komponenta.\u003C\u002Fstrong> AI sistem može da pretražuje mnoge izvore, ali samo neki izvori su autoritativni za dato pitanje. Ispravan izvor može biti aplikaciona baza podataka za trenutno stanje naloga, odobreni dokument politike za pravilo, izvorni kod za implementirano ponašanje, zvanična specifikacija za protokol ili primarni istraživački artefakt za uočenu tvrdnju. Dobra arhitektura čini taj autoritet eksplicitnim umesto da traži od modela da ga zaključi na osnovu relevantnosti.","Direktan odgovor","info","callout",{},{"id":229,"data":230,"type":226,"tunes":234},"terminology",{"body":231,"title":232,"variant":233},"„Izvor istine“ se široko koristi u softveru, podacima i organizacionoj praksi, ali nije standardizovana AI komponenta sa jednom univerzalnom formalnom definicijom. Ovaj članak ga koristi kao arhitektonski koncept: \u003Cstrong>izvor ili mapiranje autoriteta koje određuje koji dokazi smeju da utvrde određenu činjenicu ili stanje\u003C\u002Fstrong>. Standardi provenijencije kao što je W3C PROV opisuju poreklo i izvođenje; oni automatski ne odlučuju koji izvor vaša aplikacija treba da tretira kao autoritativan.","Terminološka granica","note",{},{"id":236,"data":237,"type":226,"tunes":240},"current",{"body":238,"title":239,"variant":233},"Osnovna razlika između autoriteta, provenijencije, preuzimanja i konteksta modela je stabilna. NIST AI RMF 1.0 se trenutno revidira, ali trenutni NIST Playbook i dalje eksplicitno preporučuje dokumentovanje provenijencije podataka, uključujući izvore, poreklo, transformacije, zavisnosti, ograničenja i metapodatke. W3C PROV ostaje W3C preporuka za interoperabilne opise provenijencije.","Napomena o trenutnom izvoru — 8. oktobar 2026.",{},{"id":242,"data":243,"type":247,"tunes":248},"toc",{"title":244,"maxLevel":245,"minLevel":246},"Sadržaj",3,2,"tableOfContents",{},{"id":250,"data":251,"type":42,"tunes":253},"h-meaning",{"text":252,"level":246},"Šta „Izvor istine“ zaista znači",{},{"id":255,"data":256,"type":218,"tunes":258},"p-meaning-1",{"text":257},"Fraza se često pogrešno razume kao „jedna baza podataka koja sadrži sve.“ To može biti tačno u uskom sistemu, ali je obično previše pojednostavljeno za AI. Prava AI aplikacija može kombinovati operativne baze podataka, dokumente, API-je, vektorske indekse, korisnički unos, memoriju modela, eksterne web izvore i generisane sažetke.",{},{"id":260,"data":261,"type":218,"tunes":263},"p-meaning-2",{"text":262},"Ti izvori nemaju jednak autoritet. Priručnik za korisničku podršku može definisati politiku, ali ne trenutno stanje korisnika. CRM može definisati trenutnog vlasnika naloga, ali ne pravno značenje propisa. Repozitorijum izvornog koda može definisati implementirano ponašanje, dok specifikacija proizvoda definiše nameravano ponašanje. Arhitektura stoga mora odgovoriti na preciznije pitanje: koji izvor je autoritativan za ovu konkretnu tvrdnju?",{},{"id":265,"data":266,"type":218,"tunes":268},"p-meaning-3",{"text":267},"Ovo čini Izvor istine odnosom između tvrdnje i autoriteta, a ne samo svojstvom tehnologije skladištenja.",{},{"id":270,"data":271,"type":226,"tunes":275},"principle",{"body":272,"title":273,"variant":274},"\u003Cstrong>Jedan sistem može imati mnogo Izvora istine jer je autoritet specifičan za činjenicu.\u003C\u002Fstrong> Važno arhitektonsko svojstvo nije da jedna baza podataka globalno pobeđuje, već da svaka važna činjenica ili stanje ima eksplicitnog autoritativnog vlasnika.","Osnovni princip","success",{},{"id":277,"data":278,"type":42,"tunes":280},"h-simple",{"text":279,"level":246},"Najjednostavniji primer",{},{"id":282,"data":283,"type":218,"tunes":285},"p-simple-1",{"text":284},"Korisnik pita AI asistenta: „Koji je moj trenutni plan pretplate?“ Asistent ima tri moguća ulaza: transkript podrške od prošlog meseca, indeksirani dokument centra za pomoć koji opisuje tipove planova i bazu podataka za naplatu uživo.",{},{"id":287,"data":288,"type":218,"tunes":290},"p-simple-2",{"text":289},"Transkript podrške može pomenuti da je korisnik imao Pro plan. Dokument centra za pomoć objašnjava šta Pro znači. Ali zapis o naplati uživo je autoritativni izvor za trenutno stanje korisnikove pretplate.",{},{"id":292,"data":293,"type":218,"tunes":295},"p-simple-3",{"text":294},"Semantički pretraživač bi mogao rangirati transkript podrške iznad zapisa o naplati jer sadrži jezik bliži pitanju. To rangiranje ipak ne bi učinilo transkript autoritativnim. Relevantnost i autoritet su različite dimenzije.",{},{"id":297,"data":298,"type":329,"tunes":330},"simple-comparison",{"rows":299,"title":317,"layout":318,"columns":319},[300,305,309,313],{"id":301,"label":302,"values":303},"billing","Baza podataka za naplatu",[304,304,304],"",{"id":306,"label":307,"values":308},"help","Dokumentacija centra za pomoć",[304,304,304],{"id":310,"label":311,"values":312},"transcript","Stari transkript podrške",[304,304,304],{"id":314,"label":315,"values":316},"model","Memorija modela",[304,304,304],"Isto pitanje može uključivati različite uloge izvora","table",[320,323,326],{"id":321,"label":322},"source","Izvor",{"id":324,"label":325},"role","Uloga",{"id":327,"label":328},"authority","Autoritet za trenutni plan?","comparison",{},{"id":332,"data":333,"type":42,"tunes":335},"h-stops",{"text":334,"level":246},"Gde se jednostavan primer zaustavlja",{},{"id":337,"data":338,"type":218,"tunes":340},"p-stops-1",{"text":339},"Nema svaki domen jedan neosporni autoritet. Istorijsko istraživanje može sadržati konfliktne primarne izvore. Naučne tvrdnje mogu se razvijati kako se pojavljuju nove studije. Pravno tumačenje može zavisiti od jurisdikcije, datuma i autoriteta suda. Ponašanje proizvoda može se razlikovati između dokumentacije i implementiranog koda.",{},{"id":342,"data":343,"type":218,"tunes":345},"p-stops-2",{"text":344},"U tim slučajevima ispravna arhitektura nije da se izmisli jedan pobednik. Već da se očuvaju konkurentski izvori, njihova provenijencija, njihova klasa autoriteta, njihov primenljivi obim i nerešena kontradikcija. Pouzdan sistem Izvora istine mora biti sposoban da predstavi nesigurnost i neslaganje.",{},{"id":347,"data":348,"type":42,"tunes":350},"h-authority",{"text":349,"level":246},"Nadležnost je ograničena tvrdnjom, verzijom i vremenom",{},{"id":352,"data":353,"type":318,"tunes":387},"authority-table",{"content":354,"stretched":43,"withHeadings":14},[355,359,363,367,371,375,379,383],[356,357,358],"Pitanje","Mogući merodavni izvor","Zašto je opseg važan",[360,361,362],"Koje je trenutno stanje na korisničkom računu?","Knjiga \u002F računovodstveni sistem evidencije","Istorijski izvozi mogu biti tačni za ranije vreme, ali ne i za trenutno stanje.",[364,365,366],"Šta trenutno dozvoljava politika kompanije?","Odobrena trenutna verzija politike","Starija politika može ostati validan dokaz o prošlim pravilima, ali ne i o sadašnjim.",[368,369,370],"Koji je kod zaista raspoređen?","Artefakt raspoređivanja \u002F commit \u002F zapis o izdanju","Glavna grana se može razlikovati od produkcije.",[372,373,374],"Šta je ugovor navodio u trenutku potpisivanja?","Izvršena verzija ugovora","Nacrt ili kasniji šablon nije merodavan za potpisani sporazum.",[376,377,378],"Šta specificira tehnički protokol?","Trenutna zvanična specifikacija za relevantnu verziju","Objašnjenje na blogu može biti korisno, ali je sekundarni dokaz.",[380,381,382],"Šta se dogodilo u istorijskom događaju?","Relevantni primarni dokazi plus eksplicitna kritika izvora","Možda ne postoji jedinstveni autoritet; suprotstavljeni dokazi moraju ostati vidljivi.",[384,385,386],"Šta korisnik preferira?","Trenutna eksplicitna korisnička postavka ili potvrđena preferencija","Staro pamćenje razgovora može biti zastarelo ili zamenjeno.",{},{"id":389,"data":390,"type":218,"tunes":392},"p-authority-1",{"text":391},"Reč „istina“ stoga može biti obmanjujuća ako se ne navede njena granica. U arhitekturi, Izvor istine se obično bolje razume kao izvor ovlašćen da utvrdi određenu tvrdnju pod definisanim uslovima.",{},{"id":394,"data":395,"type":42,"tunes":397},"h-what-is-not",{"text":396,"level":246},"Šta Izvor istine nije",{},{"id":399,"data":400,"type":42,"tunes":402},"h-vs-sor",{"text":401,"level":245},"Izvor istine naspram sistema evidencije",{},{"id":404,"data":405,"type":218,"tunes":407},"p-vs-sor-1",{"text":406},"Sistem evidencije je obično merodavni operativni sistem za klasu zapisa: na primer, knjiga naplate, matični zapis ljudskih resursa ili baza podataka narudžbina. To je jedna uobičajena implementacija autoriteta Izvora istine.",{},{"id":409,"data":410,"type":218,"tunes":412},"p-vs-sor-2",{"text":411},"Ali Izvor istine je širi. Potpisani PDF ugovor, zvanični standard, artefakt raspoređivanja ili primarni arhivski dokument mogu biti merodavni, a da nisu transakcioni sistem evidencije.",{},{"id":414,"data":415,"type":42,"tunes":417},"h-vs-prov",{"text":416,"level":245},"Izvor istine naspram provenijencije",{},{"id":419,"data":420,"type":218,"tunes":422},"p-vs-prov-1",{"text":421},"Provenijencija odgovara na pitanja kao što su: Odakle potiču ovi podaci? Ko ili šta ih je proizveo? Koja transformacija je stvorila ovaj derivat? Koji prethodni entitet je korišćen? W3C PROV modeluje entitete, aktivnosti, agente i derivacije kako bi se poreklo i odgovornost mogli predstaviti.",{},{"id":424,"data":425,"type":218,"tunes":427},"p-vs-prov-2",{"text":426},"Provenijencija sama po sebi ne uspostavlja autoritet. Znati da vrednost potiče iz tabele koju je napisao određeni zaposleni pomaže u njenoj proceni, ali aplikacija i dalje zahteva pravilo koje kaže da li je ta tabela merodavna za tvrdnju.",{},{"id":429,"data":430,"type":42,"tunes":432},"h-vs-evidence",{"text":431,"level":245},"Izvor istine naspram dokaza",{},{"id":434,"data":435,"type":218,"tunes":437},"p-vs-evidence-1",{"text":436},"Dokaz podržava ili opovrgava tvrdnju. Izvor istine definiše koji izvor ima autoritet da reši ili snažno ograniči tu tvrdnju u trenutnom kontekstu aplikacije.",{},{"id":439,"data":440,"type":218,"tunes":442},"p-vs-evidence-2",{"text":441},"Izvor može biti vredan dokaz, a da ne bude merodavan. Pet korisničkih imejlova može biti dokaz da korisnici ne vole radni tok, ali oni nisu sistem evidencije za trenutnu konfiguraciju proizvoda.",{},{"id":444,"data":445,"type":42,"tunes":447},"h-vs-rag",{"text":446,"level":245},"Izvor istine naspram RAG-a",{},{"id":449,"data":450,"type":218,"tunes":452},"p-vs-rag-1",{"text":451},"RAG je obrazac pretraživanja. Pronalazi informacije i dostavlja izabrani sadržaj modelu. RAG automatski ne zna koji izvor zaslužuje autoritet.",{},{"id":454,"data":455,"type":218,"tunes":457},"p-vs-rag-2",{"text":456},"RAG tok može da pronađe zastareo dokument, sekundarni rezime ili veoma sličan ali ne-merodavan izvor. Autoritet izvora mora biti kodiran kroz dizajn korpusa, metapodatke, filtere, politiku rangiranja, validaciju ili provere nakon pretraživanja.",{},{"id":459,"data":460,"type":42,"tunes":462},"h-vs-vector",{"text":461,"level":245},"Izvor istine naspram vektorske baze podataka",{},{"id":464,"data":465,"type":218,"tunes":467},"p-vs-vector-1",{"text":466},"Vektorska baza podataka čuva ili indeksira reprezentacije koje se koriste za semantičko pretraživanje. To je sloj pristupa, a ne automatski sloj istine.",{},{"id":469,"data":470,"type":218,"tunes":472},"p-vs-vector-2",{"text":471},"Isti autoritativni dokument može biti deljen u delove, ugrađen, kopiran i ponovo indeksiran mnogo puta. Vektorski zapis treba da zadrži referencu nazad na autoritativni izvor i verziju, umesto da postane novi autoritet bez mogućnosti praćenja.",{},{"id":474,"data":475,"type":42,"tunes":477},"h-vs-memory",{"text":476,"level":245},"Izvor istine naspram memorije",{},{"id":479,"data":480,"type":218,"tunes":482},"p-vs-memory-1",{"text":481},"Memorija agenta ili aplikacije čuva informacije koje mogu biti korisne kasnije. Memorija može sačuvati prethodnu odluku, preferenciju ili zapažanje, ali može postati zastarela.",{},{"id":484,"data":485,"type":218,"tunes":487},"p-vs-memory-2",{"text":486},"Za promenljivo ili posledično stanje, pouzdan agent bi normalno trebalo ponovo da pročita autoritativni trenutni izvor, umesto da pretpostavi da je zapamćeno stanje još uvek tačno.",{},{"id":489,"data":490,"type":42,"tunes":492},"h-vs-context",{"text":491,"level":245},"Izvor istine naspram konteksta",{},{"id":494,"data":495,"type":218,"tunes":497},"p-vs-context-1",{"text":496},"Kontekst je ono što model prima tokom trenutne inferencije. Autoritativne informacije mogu biti odsutne iz konteksta, dok neautoritativne informacije mogu biti prisutne.",{},{"id":499,"data":500,"type":218,"tunes":502},"p-vs-context-2",{"text":501},"Izgradnja konteksta stoga zahteva politiku svesnu autoriteta: preuzeti ili pročitati izvor koji je ovlašćen da definiše tvrdnju, zatim sačuvati dovoljno metapodataka da model ili validator mogu da razumeju njen obim.",{},{"id":504,"data":505,"type":42,"tunes":507},"h-vs-ground-truth",{"text":506,"level":245},"Izvor istine naspram evaluacione osnovne istine",{},{"id":509,"data":510,"type":218,"tunes":512},"p-vs-ground-1",{"text":511},"Evaluaciona osnovna istina je referentni odgovor, oznaka ili ishod prema kojem se sistem ocenjuje. Može se izvesti iz autoritativnih izvora, stručnog presuđivanja ili pripremljenih test podataka.",{},{"id":514,"data":515,"type":218,"tunes":517},"p-vs-ground-2",{"text":516},"Osnovna istina je stoga evaluacioni konstrukt. Izvor istine je konstrukt autoriteta aplikacije\u002Fdomena. Mogu se preklapati, ali nisu zamenljivi.",{},{"id":519,"data":520,"type":42,"tunes":522},"h-vs-quality",{"text":521,"level":245},"Izvor istine naspram kvaliteta podataka",{},{"id":524,"data":525,"type":218,"tunes":527},"p-vs-quality-1",{"text":526},"Autoritativni izvor i dalje može sadržati greške. Autoritet govori koji izvor zvanično upravlja činjenicom; kvalitet podataka pita da li je taj izvor tačan, potpun, pravovremen, konzistentan i prikladan za svrhu.",{},{"id":529,"data":530,"type":218,"tunes":532},"p-vs-quality-2",{"text":531},"Kada se zna da je autoritativni sistem pogrešan, arhitektura treba da zabeleži defekt, proces ispravke ili izuzetak, umesto da tiho zameni nezvaničnim izvorom i sakrije neslaganje.",{},{"id":534,"data":535,"type":42,"tunes":537},"h-model",{"text":536,"level":246},"Praktičan arhitektonski model izvora istine",{},{"id":539,"data":540,"type":226,"tunes":543},"model-note",{"body":541,"title":542,"variant":233},"Model ispod je praktična sinteza za AI sisteme. Nije W3C, NIST ili ISO standard. Razdvaja autoritet, poreklo, preuzimanje i kontekst okrenut modelu, jer se te odgovornosti često pogrešno spajaju.","Predloženi arhitektonski model",{},{"id":545,"data":546,"type":574,"tunes":575},"sot-flow",{"steps":547,"title":572,"orientation":573},[548,551,554,557,560,563,566,569],{"label":549,"description":550},"1. Definiši tip tvrdnje","Identifikuj šta korisnik zapravo pita: trenutno stanje, politiku, istorijsku činjenicu, tehničku specifikaciju, korisničku preferenciju, izračunavanje ili interpretaciju.",{"label":552,"description":553},"2. Razreši autoritet","Utvrdi koji izvor ili klasa autoriteta je ovlašćena da definiše taj tip tvrdnje za zahtevani obim, verziju i vreme.",{"label":555,"description":556},"3. Pribavi dokaze","Pročitaj ili preuzmi autoritativni izvor i sve potrebne prateće ili suprotstavljene dokaze.",{"label":558,"description":559},"4. Sačuvaj poreklo","Prenesi identitet izvora, verziju, vremensku oznaku, lokator, istoriju transformacija i metapodatke o odgovornosti.",{"label":561,"description":562},"5. Izgradi kontekst modela","Dostavi relevantne dokaze modelu bez odbacivanja metapodataka o autoritetu i primenljivosti.",{"label":564,"description":565},"6. Generiši ili izračunaj","Model može da sumira, uporedi, zaključuje ili transformiše dokaze, ali ne nasleđuje autoritet izvora samo obradom.",{"label":567,"description":568},"7. Validiraj tvrdnju","Proveri da je odgovor podržan od strane pravog izvora i da ostaje unutar svog obima i granice važenja.",{"label":570,"description":571},"8. Sačuvaj kontradikcije","Ako se autoritativni ili relevantni izvori ne slažu, izloži sukob umesto da izmišljaš lažnu sigurnost.","Put AI odgovora svestan autoriteta","auto","processFlow",{},{"id":577,"data":578,"type":42,"tunes":580},"h-registry",{"text":579,"level":246},"Autoritet treba da bude eksplicitan, a ne izveden iz sličnosti",{},{"id":582,"data":583,"type":218,"tunes":585},"p-registry-1",{"text":584},"Jedan robustan obrazac implementacije je registar autoriteta ili ekvivalentan sloj politike koji mapira klase tvrdnji na klase autoritativnih izvora. Implementacija može biti kod, metapodaci, konfiguracija ili pravila domena; važno svojstvo je da je autoritet nameran.",{},{"id":587,"data":588,"type":318,"tunes":618},"registry-table",{"content":589,"stretched":43,"withHeadings":14},[590,594,598,602,606,610,614],[591,592,593],"Klasa tvrdnje","Pravilo autoriteta","Rezervno ponašanje",[595,596,597],"Trenutno stanje računa","Čitanje usluge aktivnog računa \u002F sistema evidencije","Ako nije dostupno, prijaviti da trenutno stanje ne može biti verifikovano.",[599,600,601],"Dokumentacija proizvoda","Trenutna odobrena verzija dokumentacije","Starija verzija može biti prikazana samo uz upozorenje o verziji.",[603,604,605],"Implementirano ponašanje softvera","Relevantno objavljeno izdanje \u002F izvorni artefakt","Sama dokumentacija ne može dokazati objavljeno ponašanje.",[607,608,609],"Interna politika","Odobreni repozitorijum politika i aktivna verzija","Nacrti su prateći materijal, a ne trenutni autoritet.",[611,612,613],"Spoljni tehnički standard","Publikacija zvaničnog standardizacionog tela za relevantnu verziju","Sekundarna objašnjenja mogu pojasniti, ali ne mogu nadjačati specifikaciju.",[615,616,617],"Istraživačka tvrdnja","Politika dokaza primjerena domenu","Sačuvati suprotstavljene dokaze i pouzdanost umjesto forsiranja jednog izvora.",{},{"id":620,"data":621,"type":42,"tunes":623},"h-retrieval",{"text":622,"level":246},"Pretraga treba da koristi autoritet kao ograničenje za rangiranje",{},{"id":625,"data":626,"type":218,"tunes":628},"p-retrieval-1",{"text":627},"Semantička relevantnost odgovara na pitanje „Koji kandidat izgleda povezano s ovim upitom?“ Autoritet odgovara na pitanje „Koji kandidat smije da utvrdi ovu činjenicu?“ Produkcioni sistem pretrage često zahtijeva oboje.",{},{"id":630,"data":631,"type":218,"tunes":633},"p-retrieval-2",{"text":632},"Korisna sekvenca je prvo ograničiti prostor kandidata prema identitetu, zakupcu, klasi izvora, statusu, verziji ili datumu, a zatim rangirati relevantne dokaze unutar dozvoljenog prostora. Ako se relevantnost izračunava prije kritičnih autorizacijskih ili autoritetskih filtera, pipeline može vratiti uvjerljiv, ali nevažeći rezultat.",{},{"id":635,"data":636,"type":226,"tunes":640},"relevance-authority",{"body":637,"title":638,"variant":639},"Pretraživač može rangirati zastarjeli sažetak iznad trenutnog primarnog izvora. Jezički model može preferirati bolje napisano objašnjenje umjesto zvaničnog zapisa. Nijedno ponašanje ne mijenja koji izvor posjeduje činjenicu.","Najveća sličnost nije najveći autoritet","warning",{},{"id":642,"data":643,"type":42,"tunes":645},"h-freshness",{"text":644,"level":246},"Svježina je dio autoriteta",{},{"id":647,"data":648,"type":218,"tunes":650},"p-freshness-1",{"text":649},"Mnogi kvarovi izvora istine su zapravo vremenski kvarovi. Ispravan izvor je bio poznat, ali sistem je koristio stari snimak, zastarjeli embedding, keširani API odgovor ili zamijenjeni dokument.",{},{"id":652,"data":653,"type":218,"tunes":655},"p-freshness-2",{"text":654},"Pravilo autoriteta stoga treba uključivati semantiku poništavanja ili osvježavanja tamo gdje se činjenica može promijeniti. „CRM je autoritativan“ je nepotpuno kada aplikacija čita sedmicu star replicirani izvoz.",{},{"id":657,"data":658,"type":42,"tunes":660},"h-derivation",{"text":659,"level":246},"Izvedene vrijednosti zahtijevaju trag natrag do autoritativnih ulaza",{},{"id":662,"data":663,"type":218,"tunes":665},"p-derivation-1",{"text":664},"Neke važne činjenice nisu pohranjene direktno. One se izračunavaju iz autoritativnih ulaza: rezultat rizika, ukupan iznos računa, status podobnosti ili agregirana metrika.",{},{"id":667,"data":668,"type":218,"tunes":670},"p-derivation-2",{"text":669},"Za izvedene vrijednosti, arhitektura izvora istine treba da sačuva ulazne autoritete, verziju transformacije ili izračuna i vrijeme izvršenja. W3C PROV razlikovanje između entiteta, aktivnosti i derivacija je ovdje korisno jer modeluje kako je jedan entitet proizveden iz drugih.",{},{"id":672,"data":673,"type":226,"tunes":676},"derived-rule",{"body":674,"title":675,"variant":274},"Izračunata vrijednost je samo onoliko aktuelna i autoritativna koliko su njeni ulazi, transformacija i uslovi važenja. Sačuvajte lozu potrebnu za ponovni izračun ili reviziju.","Izvedena istina je uslovna",{},{"id":678,"data":679,"type":42,"tunes":681},"h-conflict",{"text":680,"level":246},"Šta se dešava kada se autoritativni izvori ne slažu?",{},{"id":683,"data":684,"type":218,"tunes":686},"p-conflict-1",{"text":685},"Sukobi nisu rubni slučajevi u ozbiljnim sistemima znanja. Potpisan ugovor može se ne slagati s poljem u CRM-u. Produkciono ponašanje može se ne slagati s dokumentacijom. Dva primarna istorijska izvora mogu protivrječiti jedan drugom. Trenutna politika može biti u sukobu sa zastarjelom lokalnom kopijom.",{},{"id":688,"data":689,"type":218,"tunes":691},"p-conflict-2",{"text":690},"Sistem zahtijeva politiku rješavanja primjerenu domenu. Ponekad jedan autoritet jasno nadjačava drugi. Ponekad novija verzija zamjenjuje stariju. Ponekad stručnjak ili vlasnik poslovanja mora presuditi. A ponekad je ispravan rezultat jednostavno: dokazi su neriješeni.",{},{"id":693,"data":694,"type":318,"tunes":717},"conflict-table",{"content":695,"stretched":43,"withHeadings":14},[696,699,702,705,708,711,714],[697,698],"Tip sukoba","Tipično postupanje",[700,701],"Trenutna vs zamijenjena verzija","Koristiti trenutnu verziju za sadašnje stanje; zadržati stariju verziju kao istorijski dokaz.",[703,704],"Sistem evidencije vs zastarjela replika","Koristiti sistem evidencije; označiti problem svježine replikacije.",[706,707],"Ugovor vs CRM transkripcija","Izvršeni ugovor upravlja ugovornim tekstom; CRM neslaganje postaje zadatak ispravke.",[709,710],"Dokumentacija vs objavljeno ponašanje","Razlikovati namjeravano ponašanje od posmatranog\u002Fobjavljenog ponašanja; ne spajati ih tiho.",[712,713],"Dva vjerodostojna primarna izvora","Sačuvati oba, procijeniti porijeklo i obim, i predstaviti neriješeno neslaganje ako ne postoji nadležni autoritet.",[715,716],"Korisničko pamćenje vs trenutna korisnička postavka","Koristiti trenutnu eksplicitnu postavku; označiti pamćenje kao zamijenjeno gdje je prikladno.",{},{"id":719,"data":720,"type":42,"tunes":722},"h-web",{"text":721,"level":246},"Web pretraga je otkrivanje, ne automatski dokaz",{},{"id":724,"data":725,"type":218,"tunes":727},"p-web-1",{"text":726},"Pretraživači su odlični sistemi za otkrivanje. Isječci pretrage, rangiranje rezultata i generisani sažeci nisu automatski primarni dokaz.",{},{"id":729,"data":730,"type":218,"tunes":732},"p-web-2",{"text":731},"Za tvrdnje koje zahtevaju autoritet, rezultat pretrage treba da vodi do originalne publikacije, zvaničnog zapisa, izvornog dokumenta, skupa podataka ili drugog odgovarajućeg artefakta. Stranica rezultata pomaže u lociranju izvora; ona ne nasleđuje autoritet izvora.",{},{"id":734,"data":735,"type":42,"tunes":737},"h-ai",{"text":736,"level":246},"Jezički model ne treba sam da odlučuje o autoritetu",{},{"id":739,"data":740,"type":218,"tunes":742},"p-ai-1",{"text":741},"Model može pomoći u klasifikaciji pitanja, izdvajanju tvrdnji ili upoređivanju dokaza, ali autoritet ne treba da zavisi samo od preferencije modela. Modeli optimizuju generisanje iz konteksta; oni ne poseduju garantovani domen-specifični registar koji određuje koja baza podataka, dokument ili organizacija poseduje svaku činjenicu.",{},{"id":744,"data":745,"type":218,"tunes":747},"p-ai-2",{"text":746},"Zbog toga arhitektura aplikacije treba da kodira kritična pravila autoriteta deterministički gde je to praktično. Model može da rezonuje unutar granice, ali sama granica ne treba da se iznova stvara za svaki prompt.",{},{"id":749,"data":750,"type":42,"tunes":752},"h-observability",{"text":751,"level":246},"Autoritet mora da preživi trag izvršavanja",{},{"id":754,"data":755,"type":218,"tunes":757},"p-observability-1",{"text":756},"Ako je odgovor u produkciji dovoljno važan za reviziju, trag treba da omogući rekonstrukciju toga koji su izvori konsultovani, koja verzija je korišćena, koji odlomak ili zapis je podržao tvrdnju, koje transformacije su se dogodile i da li su bili dostupni suprotstavljeni dokazi.",{},{"id":759,"data":760,"type":218,"tunes":762},"p-observability-2",{"text":761},"Ovo je u skladu sa širim principom provenijencije u W3C PROV i sa smernicama NIST AI RMF Playbook za dokumentovanje izvora, porekla, transformacija, zavisnosti, ograničenja i metapodataka.",{},{"id":764,"data":765,"type":42,"tunes":767},"h-implementation",{"text":766,"level":246},"Dokaz originalne implementacije: Source of Truth Research Engine",{},{"id":769,"data":770,"type":226,"tunes":773},"implementation-note",{"body":771,"title":772,"variant":233},"Source of Truth Research Engine je moj sopstveni dokaz implementacije za arhitekturu koja stavlja dokaze na prvo mesto. On demonstrira jedan konkretan način da se razdvoje otkrivanje, pribavljanje izvora, provenijencija, tvrdnje, klase dokaza, kontradikcije i zaključci. To je obrazac implementacije, a ne univerzalni industrijski standard.","Dokaz originalne implementacije",{},{"id":775,"data":776,"type":218,"tunes":778},"p-impl-1",{"text":777},"Engine je dizajniran oko sledivog pipeline-a, a ne direktnog AI sumiranja: istraživački zadatak → pretraga → originalni izvor ili digitalni artefakt → lokalni snimak → SHA-256 → ID izvora → tvrdnja → klasa dokaza → relacija ili kontradikcija → interpretacija → zaključak.",{},{"id":780,"data":781,"type":218,"tunes":783},"p-impl-2",{"text":782},"Njegovo zajedničko jezgro dokaza čuva Sources, Artifacts, provenijenciju, Claims, Relations, Contradictions, Reference Model i revizorski trag. Različiti režimi istraživanja mogu da dele to jezgro dok primenjuju različite domenske metodologije.",{},{"id":785,"data":786,"type":218,"tunes":788},"p-impl-3",{"text":787},"Arhitektura namerno razdvaja otkrivanje od dokaza. Isječci pretrage se ne tretiraju kao dokaz, nazivi datoteka se ne tretiraju kao sadržaj, AI sažeci se ne tretiraju kao primarni izvori, a semantička sličnost je samo signal za otkrivanje dok se rezultat ne poveže nazad sa konkretnim izvorom i lokatorom.",{},{"id":790,"data":791,"type":218,"tunes":793},"p-impl-4",{"text":792},"Originalne datoteke se čuvaju, a lokalni bajtovi dobijaju SHA-256 identifikatore. Kontradikcije i odbačene hipoteze se ne brišu tiho. Novim dokazima je dozvoljeno da promene trenutni referentni model, dok prethodni put dokaza ostaje podložan reviziji.",{},{"id":795,"data":796,"type":318,"tunes":825},"implementation-table",{"content":797,"stretched":43,"withHeadings":14},[798,801,804,807,810,813,816,819,822],[799,800],"Implementirano pravilo","Zašto je važno za arhitekturu izvora istine",[802,803],"Pretraga ≠ dokaz","Rangiranje otkrivanja ne može tiho da postane autoritet.",[805,806],"Naziv datoteke ≠ sadržaj","Tragovi u metapodacima ne mogu da zamene čitanje stvarnog artefakta.",[808,809],"Lokalni snimak + SHA-256","Dokaz može da se veže za tačne bajtove umesto za promenljivu udaljenu oznaku.",[811,812],"ID izvora + tačan lokator","Tvrdnje mogu da se prate nazad do konkretne lokacije dokaza.",[814,815],"Razdvajanje tvrdnje i dokaza","Tvrdnja se ne meša sa materijalom koji je podržava.",[817,818],"Kontradikcije sačuvane","Sistem može da predstavi nerešeno neslaganje umesto da prepiše istoriju.",[820,821],"Semantička sličnost je samo otkrivanje","Relevantnost pretrage je eksplicitno odvojena od dokaznog autoriteta.",[823,824],"Novi dokazi mogu da ažuriraju model","Stanje izvora istine je verzionisano i podložno reviziji, a ne tretira se kao nepromenljiva dogma.",{},{"id":827,"data":828,"type":42,"tunes":830},"h-document-engine",{"text":829,"level":246},"Aaasaasa Document & Knowledge Engine: primena iste granice na poslovne dokumente",{},{"id":832,"data":833,"type":218,"tunes":835},"p-doc-1",{"text":834},"Koncept Aaasaasa Document & Knowledge Engine proširuje isti princip dizajna na poslovnu dokumentaciju: korisnici treba da mogu da pretražuju dokumente, postavljaju pitanja zasnovana na izvorima i pregledaju kolekcije prema eksplicitnim kriterijumima, uz očuvanje razlike između onoga što dokument navodi i onoga što sistem zaključuje.",{},{"id":837,"data":838,"type":218,"tunes":840},"p-doc-2",{"text":839},"Važno arhitektonsko pravilo je da univerzalno jezgro za pretragu ne čini svaku kolekciju jednako autoritativnom. Ugovorni dokumenti, evidencija održavanja, finansijski dokumenti i istraživački materijal zahtevaju različita pravila autoriteta, validacije i pokrivenosti čak i kada dele infrastrukturu za unos i pretragu.",{},{"id":842,"data":843,"type":42,"tunes":845},"h-failures",{"text":844,"level":246},"Uobičajeni načini neuspeha izvora istine",{},{"id":847,"data":848,"type":318,"tunes":889},"failures-table",{"content":849,"stretched":43,"withHeadings":14},[850,853,856,859,862,865,868,871,874,877,880,883,886],[851,852],"Način neuspeha","Šta ide naopako",[854,855],"Model se tretira kao izvor istine","Parametarsko znanje može biti zastarelo, nepotpuno, neproverljivo ili izvan autoritativnog domena aplikacije.",[857,858],"Najbolji rezultat pretrage automatski pobeđuje","Sličnost se pogrešno smatra autoritetom.",[860,861],"Vektorska baza podataka postaje autoritativna","Izvedeni zapisi indeksa gube identitet i verziju originalnog izvora.",[863,864],"Sve se kopira u jednu bazu znanja","Kopije prikrivaju vlasništvo, svežinu i puteve ispravke.",[866,867],"Memorija se ponovo koristi kao trenutno stanje","Stara zapažanja tiho nadjačavaju trenutni sistem evidencije.",[869,870],"Nema metapodataka o verziji","Ispravan dokument se koristi za pogrešan vremenski period.",[872,873],"Nema lokatora izvora","Citiranje postoji, ali prateći odlomak ili zapis ne može se proveriti.",[875,876],"Konflikti se prepisuju","Sistem izgleda konzistentno uništavanjem dokaza o neslaganju.",[878,879],"Generisani rezimei zamenjuju originale","Transformacija sa gubicima postaje prividni autoritet.",[881,882],"Autoritet je globalan umesto specifičan za tvrdnju","Jednom izvoru se veruje izvan domena ili klase činjenica koje zaista poseduje.",[884,885],"Web isečak se tretira kao dokaz","Metapodaci za otkrivanje zamenjuju originalnu publikaciju.",[887,888],"Autoritativni podaci su pogrešni, ali izuzeci su skriveni","Operativni defekti postaju nevidljivi i ne mogu se transparentno ispraviti.",{},{"id":891,"data":892,"type":42,"tunes":894},"h-decision",{"text":893,"level":246},"Praktični okvir za odlučivanje o izvoru istine",{},{"id":896,"data":897,"type":574,"tunes":927},"decision-flow",{"steps":898,"title":926,"orientation":573},[899,902,905,908,911,914,917,920,923],{"label":900,"description":901},"1. Precizno navedite tvrdnju","Razdvojite trenutno stanje, istorijsko stanje, politiku, interpretaciju, predviđanje i izvedeni proračun.",{"label":903,"description":904},"2. Identifikujte vlasnika autoriteta","Odredite sistem, dokument, instituciju, osobu ili klasu dokaza odgovornu za taj tip tvrdnje.",{"label":906,"description":907},"3. Definišite obim","Navedite zakupca, jurisdikciju, proizvod, okruženje, korisnika, skup dokumenata ili drugu granicu primenljivosti.",{"label":909,"description":910},"4. Definišite vreme i verziju","Odredite da li tvrdnja zahteva trenutno stanje, istorijski snimak ili određenu verziju standarda\u002Fizdanja.",{"label":912,"description":913},"5. Sačuvajte provenijenciju","Zabeležite identitet izvora, poreklo, lokator, transformacije i odgovorne agente ili procese.",{"label":915,"description":916},"6. Definišite put pretrage\u002Fpristupa","Uverite se da aplikacija zaista može da dobije autoritativnu informaciju pod ispravnim identitetom i dozvolama.",{"label":918,"description":919},"7. Definišite politiku konflikta","Odlučite o prvenstvu, zameni, presuđivanju ili eksplicitnom ponašanju u nerazrešenom stanju.",{"label":921,"description":922},"8. Definišite poništavanje","Navedite kada se keširani, indeksirani, zapamćeni ili izvedeni prikazi moraju osvežiti.",{"label":924,"description":925},"9. Validirajte put odgovora","Proverite da se važne generisane tvrdnje mogu pratiti do predviđenog autoriteta, a ne samo do verodostojnog izvora.","Kako odlučiti šta treba da definiše tvrdnju",{},{"id":929,"data":930,"type":42,"tunes":932},"h-checklist",{"text":931,"level":246},"Kontrolna lista arhitekture izvora istine",{},{"id":934,"data":935,"type":318,"tunes":975},"checklist-table",{"content":936,"stretched":43,"withHeadings":14},[937,939,942,945,948,951,954,957,960,963,966,969,972],[356,938],"Očekivani odgovor",[940,941],"Koja tačna činjenica ili stanje se utvrđuje?","Tvrdnja dovoljno precizna da joj se dodeli autoritet.",[943,944],"Ko ili šta poseduje tu činjenicu?","Imenovani autoritativni sistem, klasa izvora ili pravilo presuđivanja.",[946,947],"Da li je autoritet aktuelan za ovaj obim?","Granica zakupca, jurisdikcije, okruženja, korisnika ili domena je eksplicitna.",[949,950],"Da li su verzija\u002Fvreme ispravni?","Trenutna, istorijska ili verzijski specifična primenljivost je poznata.",[952,953],"Može li se izvor proveriti?","Postoji stabilni identifikator, lokator ili referenca zapisa.",[955,956],"Da li je provenijencija sačuvana?","Metapodaci o poreklu, transformaciji i odgovornosti preživljavaju unos i pretragu.",[958,959],"Može li pretraga vratiti neautoritativni materijal?","Ako da, filteri ili validacija razlikuju relevantnost od autoriteta.",[961,962],"Može li se izvor promeniti?","Postoje pravila za osvežavanje, poništavanje ili zamenu.",[964,965],"Mogu li se izvori ne slagati?","Ponašanje pri konfliktu i presuđivanju je eksplicitno.",[967,968],"Može li memorija postati zastarela?","Nestabilno stanje se ponovo čita iz trenutnog autoriteta pre posledične upotrebe.",[970,971],"Može li se izvedeni odgovor reprodukovati?","Ulazi, verzija transformacije i uslovi izvršavanja su sledljivi.",[973,974],"Može li revizor rekonstruisati odgovor?","Dokazi o izvršavanju čuvaju put izvora za važne tvrdnje.",{},{"id":977,"data":978,"type":42,"tunes":980},"h-misconceptions",{"text":979,"level":246},"Uobičajene zablude",{},{"id":982,"data":983,"type":318,"tunes":1012},"misconceptions-table",{"content":984,"stretched":43,"withHeadings":14},[985,988,991,994,997,1000,1003,1006,1009],[986,987],"Zabluda","Ispravka",[989,990],"„Izvor istine znači jedna baza podataka.“","Jedna baza podataka može biti autoritativna za jedan domen; složeni sistemi obično imaju više autoriteta specifičnih za činjenice.",[992,993],"„Najnoviji dokument je automatski autoritativan.“","Novina pomaže samo kada je noviji artefakt odobren i zaista zamenjuje stariji.",[995,996],"„RAG rešava istinu.“","RAG rešava pretragu. Autoritet, provenijencija, kvalitet dokaza i validnost ostaju odvojeni problemi.",[998,999],"„Citiranje dokazuje odgovor.“","Citirani izvor mora zaista podržavati tvrdnju, imati pravi autoritet i primenjivati se na trenutni obim.",[1001,1002],"„Provenijencija nam govori šta je istinito.“","Provenijencija nam govori o poreklu i izvođenju; autoritet i ispravnost i dalje zahtevaju pravila domena i evaluaciju.",[1004,1005],"„Sistem evidencije je uvek ispravan.“","On je autoritativan za operativni zapis, ali defekti kvaliteta podataka i dalje mogu postojati i zahtevaju vidljivu ispravku.",[1007,1008],"„Ako se nekoliko izvora slaže, tvrdnja je autoritativna.“","Slaganje povećava dokaze, ali ne uspostavlja nužno vlasništvo ili primenljivost.",[1010,1011],"„AI memorija može zameniti ponovljena čitanja.“","Samo za informacije čiji je rizik zastarevanja prihvatljiv; nestabilno ili posledično stanje treba osvežiti iz autoriteta.",{},{"id":1014,"data":1015,"type":42,"tunes":1017},"h-edge",{"text":1016,"level":246},"Granični slučajevi",{},{"id":1019,"data":1020,"type":218,"tunes":1022},"p-edge-1",{"text":1021},"Neka pitanja su interpretativna, a ne činjenična. „Koja arhitektura je najbolja?“ nema jedinstveni izvor istine. Sistem može da dohvati autoritativna ograničenja i dokaze, ali konačni sud je zaključak koji treba da izloži pretpostavke i kompromise.",{},{"id":1024,"data":1025,"type":218,"tunes":1027},"p-edge-2",{"text":1026},"Neki domeni koriste distribuiran autoritet. Naučni zaključak može zavisiti od više studija, skupova podataka i replikacija. Istorijski zaključak može zavisiti od suprotstavljenih primarnih i sekundarnih dokaza. Arhitektura treba da predstavi strukturu dokaza, a ne da izmišlja centralnu bazu podataka koja navodno poseduje istinu.",{},{"id":1029,"data":1030,"type":218,"tunes":1032},"p-edge-3",{"text":1031},"Korisnik takođe može biti autoritet za subjektivne lične informacije: preferencije, ciljeve, izabrana podešavanja ili eksplicitna uputstva. Čak i tada, noviji eksplicitni unos može zameniti stariju memoriju.",{},{"id":1034,"data":1035,"type":218,"tunes":1037},"p-edge-4",{"text":1036},"Spoljni događaji mogu poništiti prethodno autoritativne podatke. Cenovni feed, sistem zaliha ili bezbednosna politika mogli su biti ispravni kada su zabeleženi, ali više nisu važeći. Provenijencija snimka čuva ono što je tada bilo istinito; ona ne čini snimak zauvek aktuelnim.",{},{"id":1039,"data":1040,"type":42,"tunes":1042},"h-limit",{"text":1041,"level":246},"Ograničenja",{},{"id":1044,"data":1045,"type":218,"tunes":1047},"p-limit-1",{"text":1046},"Arhitektura izvora istine ne može garantovati da je autoritativni izvor činjenično ispravan. Ona obezbeđuje odgovornost, provenijenciju i determinističke granice vlasništva; procesi za kvalitet podataka i verifikaciju domena ostaju neophodni.",{},{"id":1049,"data":1050,"type":218,"tunes":1052},"p-limit-2",{"text":1051},"Autoritet takođe može biti sporan. Različite institucije mogu legitimno polagati pravo na autoritet u različitim jurisdikcijama ili metodologijama. U takvim situacijama sistem treba da izloži model autoriteta i neslaganje, a ne da ih sakrije iza univerzalnog „rezultata istinitosti“.",{},{"id":1054,"data":1055,"type":218,"tunes":1057},"p-limit-3",{"text":1056},"Konačno, pravila autoriteta zahtevaju održavanje. Sistemi, vlasnici, politike, verzije i propisi se menjaju. Zastareli registar autoriteta može biti jednako opasan kao i nikakav registar.",{},{"id":1059,"data":1060,"type":42,"tunes":1062},"h-change",{"text":1061,"level":246},"Šta bi promenilo ovaj odgovor?",{},{"id":1064,"data":1065,"type":218,"tunes":1067},"p-change-1",{"text":1066},"Specifično mapiranje autoriteta menja se u zavisnosti od domena. Bankarstvo, zdravstvo, isporuka softvera, naučno istraživanje i istorijska analiza imaju različite sisteme evidencije, dokazna pravila i regulatorne obaveze.",{},{"id":1069,"data":1070,"type":218,"tunes":1072},"p-change-2",{"text":1071},"Implementacija se takođe menja sa arhitekturom. Mala aplikacija može da kodira autoritet direktno u pozivima servisa. Veća platforma može zahtevati registre, metapodatke izvora, mehanizme politika, sisteme porekla ili ugovore o podacima. Osnovni princip ostaje isti: ne dozvolite da redosled pretrage ili preferencija modela tiho odlučuju šta se smatra autoritativnim.",{},{"id":1074,"data":1075,"type":42,"tunes":1077},"h-related",{"text":1076,"level":246},"Povezano kanonsko znanje",{},{"id":1079,"data":1080,"type":218,"tunes":1082},"p-related-1",{"text":1081},"Arhitektura izvora istine je preduslov za kasnije koncepte pretrage i upravljanja jer kvalitet pretrage sam po sebi ne može da odredi da li je dokazima dozvoljeno da definišu odgovor.",{},{"id":1084,"data":1085,"type":1090,"tunes":1091},"ref-rag",{"url":1086,"title":1087,"excerpt":1088,"ctaLabel":1089},"https:\u002F\u002Fstajic.de\u002Fsr\u002Fblog\u002Fwhat-is-rag-the-simplest-explanation-of-how-it-works","Šta je RAG? Najjednostavnije objašnjenje kako funkcioniše","RAG pronalazi eksterno znanje za model. Ovaj temelj objašnjava zašto su pretraga i autoritativna istina odvojene odgovornosti.","Pročitajte osnove RAG-a","referralArticle",{},{"id":1093,"data":1094,"type":218,"tunes":1096},"p-related-2",{"text":1095},"Memorija je još jedan srodan koncept. Pouzdan agent razdvaja zapamćene informacije od trenutnog autoritativnog stanja aplikacije.",{},{"id":1098,"data":1099,"type":1090,"tunes":1104},"ref-memory",{"url":1100,"title":1101,"excerpt":1102,"ctaLabel":1103},"https:\u002F\u002Fstajic.de\u002Fsr\u002Fblog\u002Fai-agent-memory-is-not-rag-how-to-separate-memory-retrieval-state-and-context","Memorija AI agenta nije RAG: Kako razdvojiti memoriju, pretragu, stanje i kontekst","Arhitektura sa četiri sloja koja razdvaja trajnu memoriju, autoritativno stanje, pretragu i kontekst koji se stvarno dostavlja modelu.","Pročitajte članak o arhitekturi memorije",{},{"id":1106,"data":1107,"type":218,"tunes":1109},"p-related-3",{"text":1108},"Autoritet se takođe direktno povezuje sa validnošću odgovora. Čak i autoritativni izvor podržava samo tvrdnje unutar svoje verzije, datuma, obima i granice dokaza.",{},{"id":1111,"data":1112,"type":1090,"tunes":1117},"ref-avb",{"url":1113,"title":1114,"excerpt":1115,"ctaLabel":1116},"https:\u002F\u002Fstajic.de\u002Fsr\u002Fblog\u002Fthe-answer-validity-boundary-the-missing-layer-between-relevance-and-reliable-ai-answers","Granica validnosti odgovora: Sloj koji nedostaje između relevantnosti i pouzdanih AI odgovora","Okvir za eksplicitno određivanje gde AI odgovor ostaje podržan dokazima i koje promenjene okolnosti poništavaju tu podršku.","Pročitajte Granicu validnosti odgovora",{},{"id":1119,"data":1120,"type":42,"tunes":1122},"h-faq",{"text":1121,"level":246},"Često postavljana pitanja",{},{"id":1124,"data":1125,"type":1124,"tunes":1160},"faq",{"items":1126,"title":1159},[1127,1131,1135,1139,1143,1147,1151,1155],{"id":1128,"answer":1129,"question":1130},"faq1","To je autoritativni izvor ili pravilo autoriteta koje određuje koji izvor sme da utvrdi određenu činjenicu, stanje ili pravilo za definisani obim, verziju i vreme.","Šta je izvor istine u AI sistemu?",{"id":1132,"answer":1133,"question":1134},"faq2","Obično ne. Jezički model može da generiše, sumira i zaključuje, ali njegovo parametarsko znanje nije automatski autoritativno za trenutno stanje aplikacije, politiku kompanije, određenu verziju dokumenta ili činjenicu iz regulisane oblasti.","Da li je jezički model izvor istine?",{"id":1136,"answer":1137,"question":1138},"faq3","Ne automatski. Vektorska baza podataka je obično indeks ili skladište za pretragu. Trebalo bi da čuva reference na autoritativni originalni izvor i verziju, a ne da ih tiho zamenjuje.","Da li je vektorska baza podataka izvor istine za RAG?",{"id":1140,"answer":1141,"question":1142},"faq4","Poreklo opisuje odakle podaci potiču, kako su proizvedeni ili transformisani i ko ili šta je bio uključen. Pravila izvora istine određuju da li taj izvor ima autoritet za konkretnu tvrdnju.","Koja je razlika između porekla i izvora istine?",{"id":1144,"answer":1145,"question":1146},"faq5","Da. U složenim sistemima to je normalno jer različite činjenice pripadaju različitim autoritativnim sistemima ili klasama izvora.","Može li AI sistem imati više izvora istine?",{"id":1148,"answer":1149,"question":1150},"faq6","Sistemu je potrebno pravilo za rešavanje sukoba specifično za domen: prioritet, zamena verzije, stručno presuđivanje ili eksplicitno nerešeno stanje. Ne bi trebalo tiho da bira izvor koji model preferira.","Šta se dešava kada se dva autoritativna izvora ne slažu?",{"id":1152,"answer":1153,"question":1154},"faq7","Ne. RAG pronalazi kandidate. Potrebni su metapodaci svesni autoriteta, filteri, politika izvora i validacija da bi se osiguralo da važne tvrdnje koriste ispravan izvor.","Da li RAG garantuje da AI odgovor koristi izvor istine?",{"id":1156,"answer":1157,"question":1158},"faq8","Da. Autoritet i ispravnost su različita svojstva. Autoritativni sistem može sadržati defekt u kvalitetu podataka, koji treba transparentno ispraviti, a ne sakriti zamenom nezvaničnim izvorom.","Može li izvor istine biti pogrešan?","Izvor istine u AI sistemima",{},{"id":1162,"data":1163,"type":42,"tunes":1165},"h-glossary",{"text":1164,"level":246},"Pojmovnik",{},{"id":1167,"data":1168,"type":1167,"tunes":1210},"glossary",{"title":1169,"entries":1170},"Ključni pojmovi izvora istine",[1171,1175,1179,1183,1187,1190,1194,1198,1202,1206],{"term":1172,"anchor":1173,"definition":1174},"Izvor istine","source-of-truth","Autoritativni izvor ili pravilo koje je dozvoljeno da utvrdi određenu činjenicu, stanje ili pravilo za definisani obim, verziju i vreme.",{"term":1176,"anchor":1177,"definition":1178},"Sistem evidencije","system-of-record","Autoritativni operativni sistem odgovoran za definisanu klasu zapisa ili trenutno poslovno stanje.",{"term":1180,"anchor":1181,"definition":1182},"Poreklo","provenance","Informacije koje opisuju izvor, izvođenje, transformacije, odgovorne agente i istoriju podataka ili drugog entiteta.",{"term":1184,"anchor":1185,"definition":1186},"Dokaz","evidence","Informacija ili artefakt koji podržava, opovrgava ili ograničava tvrdnju.",{"term":1188,"anchor":327,"definition":1189},"Autoritet","Pravilo aplikacije ili domena koje određuje koji izvor ima pravo da definiše konkretnu tvrdnju.",{"term":1191,"anchor":1192,"definition":1193},"Svežina","freshness","Da li reprezentacija ostaje dovoljno aktuelna za tvrdnju ili operaciju u kojoj se koristi.",{"term":1195,"anchor":1196,"definition":1197},"Zamena verzije","supersession","Eksplicitna zamena starije autoritativne verzije novijom uz očuvanje istorijske sledljivosti.",{"term":1199,"anchor":1200,"definition":1201},"Osnovna istina","ground-truth","Referentni odgovor, oznaka ili ishod koji se koristi za evaluaciju sistema; to je konstrukt za evaluaciju, a ne automatski izvor istine aplikacije.",{"term":1203,"anchor":1204,"definition":1205},"Sledljivost","lineage","Trag kako podaci ili izvedene vrednosti teku i transformišu se kroz izvore i korake obrade.",{"term":1207,"anchor":1208,"definition":1209},"Granica validnosti","validity-boundary","Uslovi obima, vremena, verzije, dokaza i pretpostavki unutar kojih tvrdnja ostaje podržana.",{},{"id":1212,"data":1213,"type":42,"tunes":1215},"h-conclusion",{"text":1214,"level":246},"Zaključak",{},{"id":1217,"data":1218,"type":218,"tunes":1220},"p-conclusion-1",{"text":1219},"Pouzdan AI ne dolazi od davanja više informacija modelu. Dolazi od znanja koje informacije smeju da definišu tvrdnju, čuvanja odakle te informacije potiču, pronalaženja ispravne verzije i držanja konačnog odgovora unutar obima izvora.",{},{"id":1222,"data":1223,"type":218,"tunes":1225},"p-conclusion-2",{"text":1224},"Zato izvor istine, poreklo, pretraga, memorija i kontekst moraju ostati odvojeni koncepti. Izvor istine definiše autoritet. Poreklo objašnjava nastanak. Pretraga pronalazi kandidate. Memorija čuva izabrane prošle informacije. Kontekst je ono što model vidi. Generisanje pretvara te ulaze u izlaz.",{},{"id":1227,"data":1228,"type":218,"tunes":1230},"p-conclusion-3",{"text":1229},"Kada ti slojevi ostanu eksplicitni, AI sistem može da uradi više od zvučanja verodostojno: važne tvrdnje mogu se pratiti nazad do izvora koji je zaista imao pravo da ih utvrdi.",{},{"id":1232,"data":1233,"type":42,"tunes":1235},"h-sources",{"text":1234,"level":246},"Primarni izvori i dokazi o implementaciji",{},{"id":1237,"data":1238,"type":218,"tunes":1240},"p-sources-note",{"text":1239},"Spoljni izvori ispod podržavaju tvrdnje o poreklu i upravljanju rizikom AI. Odsek o istraživačkom motoru izvora istine je originalni dokaz o implementaciji i eksplicitno je predstavljen kao jedan obrazac implementacije, a ne kao univerzalni standard.",{},{"id":1242,"data":1243,"type":1249,"tunes":1250},"src-w3c-prov",{"link":1244,"meta":1245},"https:\u002F\u002Fwww.w3.org\u002FTR\u002Fprov-dm\u002F",{"image":1246,"title":1247,"description":1248},{"url":304},"W3C PROV-DM — PROV model podataka","W3C preporuka koja definiše model provenijencije nezavisan od domena oko entiteta, aktivnosti, agenata, izvođenja i odgovornosti.","linkTool",{},{"id":1252,"data":1253,"type":1249,"tunes":1259},"src-w3c-publications",{"link":1254,"meta":1255},"https:\u002F\u002Fwww.w3.org\u002Fgroups\u002Fwg\u002Fprov\u002Fpublications\u002F",{"image":1256,"title":1257,"description":1258},{"url":304},"W3C radna grupa za provenijenciju — publikacije","Zvanični indeks W3C PROV preporuka i povezanih specifikacija za razmenu provenijencije i ograničenja.",{},{"id":1261,"data":1262,"type":1249,"tunes":1268},"src-nist-rmf",{"link":1263,"meta":1264},"https:\u002F\u002Fwww.nist.gov\u002Fitl\u002Fai-risk-management-framework",{"image":1265,"title":1266,"description":1267},{"url":304},"NIST okvir za upravljanje rizicima AI sistema","NIST dobrovoljni okvir za uključivanje razmatranja pouzdanosti i upravljanja rizicima tokom životnog ciklusa AI sistema; AI RMF 1.0 je trenutno u fazi revizije.",{},{"id":1270,"data":1271,"type":1249,"tunes":1277},"src-nist-playbook",{"link":1272,"meta":1273},"https:\u002F\u002Fairc.nist.gov\u002Fairmf-resources\u002Fplaybook\u002F",{"image":1274,"title":1275,"description":1276},{"url":304},"NIST AI RMF priručnik","Operativne smernice usklađene sa AI RMF, uključujući prakse dokumentovanja za provenijenciju podataka, izvore, poreklo, transformacije, zavisnosti, ograničenja i metapodatke.",{},{"id":1279,"data":1280,"type":1249,"tunes":1286},"src-nist-measure",{"link":1281,"meta":1282},"https:\u002F\u002Fairc.nist.gov\u002Fairmf-resources\u002Fplaybook\u002Fmeasure\u002F",{"image":1283,"title":1284,"description":1285},{"url":304},"NIST AI RMF priručnik — merenje","Smernice o dokumentovanju merenja, provenijencije podataka i kontekstualne interpretacije izlaza AI sistema.",{},{"id":1288,"data":1289,"type":1249,"tunes":1295},"src-nist-genai",{"link":1290,"meta":1291},"https:\u002F\u002Fwww.nist.gov\u002Fpublications\u002Fartificial-intelligence-risk-management-framework-generative-artificial-intelligence",{"image":1292,"title":1293,"description":1294},{"url":304},"NIST AI 600-1 — profil generativne veštačke inteligencije","NIST profil generativne veštačke inteligencije, uključujući razmatranja provenijencije i integriteta informacija za sisteme generativne veštačke inteligencije.",{},"2.31","Izvor istine definiše koji je izvor merodavan za određenu činjenicu ili stanje. Saznajte kako se razlikuje od RAG-a, porekla, memorije, konteksta, vektorskih baza podataka i sistema evidencije.","\u002Fuploads\u002F2026\u002F10\u002Fsource-of-truth-in-ai-systems-where-reliable-knowledge-actually-comes-from-1791479103235-6bq9em.webp","source-of-truth-in-ai-systems-where-reliable-knowledge-actually-comes-from-1791479103235-6bq9em","PUBLISHED","2026-10-08T13:02:00.000Z","2026-10-08T17:02:49.770Z","2026-10-08T17:11:41.743Z",{"en":1305,"de":1306,"sr":1307,"es":1308,"fr":1309,"it":1310,"ru":1311,"zh":1312},"\u002Fblog\u002Fsource-of-truth-in-ai-systems-where-reliable-knowledge-actually-comes-from","\u002Fde\u002Fblog\u002Fsource-of-truth-in-ai-systems-where-reliable-knowledge-actually-comes-from","\u002Fsr\u002Fblog\u002Fsource-of-truth-in-ai-systems-where-reliable-knowledge-actually-comes-from","\u002Fes\u002Fblog\u002Fsource-of-truth-in-ai-systems-where-reliable-knowledge-actually-comes-from","\u002Ffr\u002Fblog\u002Fsource-of-truth-in-ai-systems-where-reliable-knowledge-actually-comes-from","\u002Fit\u002Fblog\u002Fsource-of-truth-in-ai-systems-where-reliable-knowledge-actually-comes-from","\u002Fru\u002Fblog\u002Fsource-of-truth-in-ai-systems-where-reliable-knowledge-actually-comes-from","\u002Fzh\u002Fblog\u002Fsource-of-truth-in-ai-systems-where-reliable-knowledge-actually-comes-from",[1314,1318,1322],{"id":1315,"name":1316,"slug":1317},65,"Sistem tematskog autoriteta","topical-authority",{"id":1319,"name":1320,"slug":1321},57,"Granice podataka","data-boundaries",{"id":1323,"name":1324,"slug":1325},84,"Politike i granice podataka","policy-and-data",{"id":1327,"login":1328,"email":1329,"displayName":1330},"20","rooth8233","aleksandar@stajic.de","Aleksandar Stajić",[1332,2245],{"lang":1333,"title":1334,"content":1335,"contentJson":1336,"excerpt":2244},"en","Source of Truth in AI Systems: Where Reliable Knowledge Actually Comes From","{\"time\":1791479104597,\"blocks\":[{\"id\":\"intro\",\"type\":\"paragraph\",\"data\":{\"text\":\"A Source of Truth in an AI system is the authoritative source allowed to define whether a specific fact, state or rule should be treated as true for a particular scope, version and time. It is not automatically the language model, the vector database, the top-ranked retrieved document, agent memory or the latest message in context. Reliable AI architecture must preserve which source has authority for which claim, then keep provenance, retrieval and validation connected to that authority.\"},\"tunes\":{}},{\"id\":\"direct\",\"type\":\"callout\",\"data\":{\"variant\":\"info\",\"title\":\"Direct answer\",\"body\":\"\u003Cstrong>The Source of Truth is an authority rule, not an AI component.\u003C\u002Fstrong> An AI system can search many sources, but only some sources are authoritative for a given question. The correct source may be an application database for current account state, an approved policy document for a rule, source code for implemented behavior, an official specification for a protocol, or a primary research artifact for an observed claim. Good architecture makes that authority explicit instead of asking the model to infer it from relevance.\"},\"tunes\":{}},{\"id\":\"terminology\",\"type\":\"callout\",\"data\":{\"variant\":\"note\",\"title\":\"Terminology boundary\",\"body\":\"“Source of Truth” is widely used in software, data and organizational practice, but it is not a standardized AI component with one universal formal definition. This article uses it as an architecture concept: \u003Cstrong>the source or authority mapping that determines which evidence is allowed to establish a particular fact or state\u003C\u002Fstrong>. Provenance standards such as W3C PROV describe origin and derivation; they do not automatically decide which source your application should treat as authoritative.\"},\"tunes\":{}},{\"id\":\"current\",\"type\":\"callout\",\"data\":{\"variant\":\"note\",\"title\":\"Current-source note — 8 October 2026\",\"body\":\"The core distinction between authority, provenance, retrieval and model context is stable. NIST AI RMF 1.0 is currently being revised, but the current NIST Playbook still explicitly recommends documenting data provenance, including sources, origins, transformations, dependencies, constraints and metadata. W3C PROV remains a W3C Recommendation for interoperable provenance descriptions.\"},\"tunes\":{}},{\"id\":\"toc\",\"type\":\"tableOfContents\",\"data\":{\"title\":\"Contents\",\"minLevel\":2,\"maxLevel\":3},\"tunes\":{}},{\"id\":\"h-meaning\",\"type\":\"header\",\"data\":{\"text\":\"What “Source of Truth” really means\",\"level\":2},\"tunes\":{}},{\"id\":\"p-meaning-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"The phrase is often misunderstood as “the one database that contains everything.” That can be true in a narrow system, but it is usually too simplistic for AI. A real AI application can combine operational databases, documents, APIs, vector indexes, user input, model memory, external web sources and generated summaries.\"},\"tunes\":{}},{\"id\":\"p-meaning-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"Those sources do not have equal authority. A customer-support manual may define policy but not a customer's current balance. A CRM may define the current account owner but not the legal meaning of a regulation. A source-code repository may define implemented behavior while a product specification defines intended behavior. The architecture must therefore answer a more precise question: which source is authoritative for this specific claim?\"},\"tunes\":{}},{\"id\":\"p-meaning-3\",\"type\":\"paragraph\",\"data\":{\"text\":\"This makes Source of Truth a relationship between a claim and an authority, not merely a property of a storage technology.\"},\"tunes\":{}},{\"id\":\"principle\",\"type\":\"callout\",\"data\":{\"variant\":\"success\",\"title\":\"Core principle\",\"body\":\"\u003Cstrong>One system can have many Sources of Truth because authority is fact-specific.\u003C\u002Fstrong> The important architectural property is not that one database wins globally, but that each important fact or state has an explicit authoritative owner.\"},\"tunes\":{}},{\"id\":\"h-simple\",\"type\":\"header\",\"data\":{\"text\":\"The simplest example\",\"level\":2},\"tunes\":{}},{\"id\":\"p-simple-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"A user asks an AI assistant: “What is my current subscription plan?” The assistant has three possible inputs: last month's support transcript, an indexed help-center document describing plan types, and the live billing database.\"},\"tunes\":{}},{\"id\":\"p-simple-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"The support transcript may mention that the user had a Pro plan. The help-center document explains what Pro means. But the live billing record is the authoritative source for the user's current subscription state.\"},\"tunes\":{}},{\"id\":\"p-simple-3\",\"type\":\"paragraph\",\"data\":{\"text\":\"A semantic search engine could rank the support transcript above the billing record because it contains language closer to the question. That ranking would still not make the transcript authoritative. Relevance and authority are different dimensions.\"},\"tunes\":{}},{\"id\":\"simple-comparison\",\"type\":\"comparison\",\"data\":{\"title\":\"The same question can involve different source roles\",\"layout\":\"table\",\"columns\":[{\"id\":\"source\",\"label\":\"Source\"},{\"id\":\"role\",\"label\":\"Role\"},{\"id\":\"authority\",\"label\":\"Authority for current plan?\"}],\"rows\":[{\"id\":\"billing\",\"label\":\"Billing database\",\"values\":[\"\",\"\",\"\"]},{\"id\":\"help\",\"label\":\"Help-center documentation\",\"values\":[\"\",\"\",\"\"]},{\"id\":\"transcript\",\"label\":\"Old support transcript\",\"values\":[\"\",\"\",\"\"]},{\"id\":\"model\",\"label\":\"Model memory\",\"values\":[\"\",\"\",\"\"]}]},\"tunes\":{}},{\"id\":\"h-stops\",\"type\":\"header\",\"data\":{\"text\":\"Where the simple example stops\",\"level\":2},\"tunes\":{}},{\"id\":\"p-stops-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"Not every domain has one unquestioned authority. Historical research can contain conflicting primary sources. Scientific claims can evolve as new studies appear. Legal interpretation can depend on jurisdiction, date and court authority. Product behavior can differ between documentation and deployed code.\"},\"tunes\":{}},{\"id\":\"p-stops-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"In these cases the correct architecture is not to invent a single winner. It is to preserve the competing sources, their provenance, their authority class, their applicable scope and the unresolved contradiction. A reliable Source-of-Truth system must be able to represent uncertainty and disagreement.\"},\"tunes\":{}},{\"id\":\"h-authority\",\"type\":\"header\",\"data\":{\"text\":\"Authority is scoped by claim, version and time\",\"level\":2},\"tunes\":{}},{\"id\":\"authority-table\",\"type\":\"table\",\"data\":{\"withHeadings\":true,\"stretched\":false,\"content\":[[\"Question\",\"Possible authoritative source\",\"Why scope matters\"],[\"What is the user's current account balance?\",\"Ledger \u002F accounting system of record\",\"Historical exports may be accurate for an earlier time but not current state.\"],[\"What does company policy currently allow?\",\"Approved current policy version\",\"An older policy may remain valid evidence of past rules but not present rules.\"],[\"What code is actually deployed?\",\"Deployment artifact \u002F commit \u002F release record\",\"Main branch may differ from production.\"],[\"What did a contract state when signed?\",\"Executed contract version\",\"A draft or later template is not authoritative for the signed agreement.\"],[\"What does a technical protocol specify?\",\"Current official specification for the relevant version\",\"A blog explanation may be useful but is secondary evidence.\"],[\"What happened in a historical event?\",\"Relevant primary evidence plus explicit source criticism\",\"There may be no single authority; conflicting evidence must remain visible.\"],[\"What does a user prefer?\",\"Current explicit user setting or confirmed preference\",\"Old conversation memory may be stale or superseded.\"]]},\"tunes\":{}},{\"id\":\"p-authority-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"The word “truth” can therefore be misleading unless its boundary is stated. In architecture, the Source of Truth is usually better understood as the source authorized to determine a specific proposition under defined conditions.\"},\"tunes\":{}},{\"id\":\"h-what-is-not\",\"type\":\"header\",\"data\":{\"text\":\"What a Source of Truth is not\",\"level\":2},\"tunes\":{}},{\"id\":\"h-vs-sor\",\"type\":\"header\",\"data\":{\"text\":\"Source of Truth vs system of record\",\"level\":3},\"tunes\":{}},{\"id\":\"p-vs-sor-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"A system of record is typically the authoritative operational system for a class of records: for example, a billing ledger, HR master record or order database. It is one common implementation of Source-of-Truth authority.\"},\"tunes\":{}},{\"id\":\"p-vs-sor-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"But Source of Truth is broader. A signed PDF contract, an official standard, a deployment artifact or a primary archival document may be authoritative without being a transactional system of record.\"},\"tunes\":{}},{\"id\":\"h-vs-prov\",\"type\":\"header\",\"data\":{\"text\":\"Source of Truth vs provenance\",\"level\":3},\"tunes\":{}},{\"id\":\"p-vs-prov-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"Provenance answers questions such as: Where did this data come from? Who or what produced it? Which transformation created this derivative? Which prior entity was used? W3C PROV models entities, activities, agents and derivations so origin and responsibility can be represented.\"},\"tunes\":{}},{\"id\":\"p-vs-prov-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"Provenance does not by itself establish authority. Knowing that a value came from a spreadsheet written by a specific employee helps evaluate it, but the application still needs a rule saying whether that spreadsheet is authoritative for the claim.\"},\"tunes\":{}},{\"id\":\"h-vs-evidence\",\"type\":\"header\",\"data\":{\"text\":\"Source of Truth vs evidence\",\"level\":3},\"tunes\":{}},{\"id\":\"p-vs-evidence-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"Evidence supports or contradicts a claim. A Source of Truth defines which source has the authority to settle or strongly constrain that claim in the current application context.\"},\"tunes\":{}},{\"id\":\"p-vs-evidence-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"A source can be valuable evidence without being authoritative. Five customer emails may be evidence that users dislike a workflow, but they are not the system of record for current product configuration.\"},\"tunes\":{}},{\"id\":\"h-vs-rag\",\"type\":\"header\",\"data\":{\"text\":\"Source of Truth vs RAG\",\"level\":3},\"tunes\":{}},{\"id\":\"p-vs-rag-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"RAG is a retrieval pattern. It finds information and supplies selected content to the model. RAG does not automatically know which source deserves authority.\"},\"tunes\":{}},{\"id\":\"p-vs-rag-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"A RAG pipeline can retrieve a stale document, a secondary summary or a highly similar but non-authoritative source. Source authority must be encoded through corpus design, metadata, filters, ranking policy, validation or post-retrieval checks.\"},\"tunes\":{}},{\"id\":\"h-vs-vector\",\"type\":\"header\",\"data\":{\"text\":\"Source of Truth vs vector database\",\"level\":3},\"tunes\":{}},{\"id\":\"p-vs-vector-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"A vector database stores or indexes representations used for semantic retrieval. It is an access layer, not automatically a truth layer.\"},\"tunes\":{}},{\"id\":\"p-vs-vector-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"The same authoritative document may be chunked, embedded, copied and re-indexed many times. The vector record should retain a reference back to the authoritative source and version rather than becoming an untraceable new authority.\"},\"tunes\":{}},{\"id\":\"h-vs-memory\",\"type\":\"header\",\"data\":{\"text\":\"Source of Truth vs memory\",\"level\":3},\"tunes\":{}},{\"id\":\"p-vs-memory-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"Agent or application memory stores information that may be useful later. Memory can preserve a prior decision, preference or observation, but it can become stale.\"},\"tunes\":{}},{\"id\":\"p-vs-memory-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"For volatile or consequential state, a reliable agent should normally re-read the authoritative current source rather than assume that remembered state is still true.\"},\"tunes\":{}},{\"id\":\"h-vs-context\",\"type\":\"header\",\"data\":{\"text\":\"Source of Truth vs context\",\"level\":3},\"tunes\":{}},{\"id\":\"p-vs-context-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"Context is what the model receives during the current inference. Authoritative information can be absent from context, while non-authoritative information can be present.\"},\"tunes\":{}},{\"id\":\"p-vs-context-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"Context construction therefore needs an authority-aware policy: retrieve or read the source that is allowed to define the claim, then preserve enough metadata for the model or validator to understand its scope.\"},\"tunes\":{}},{\"id\":\"h-vs-ground-truth\",\"type\":\"header\",\"data\":{\"text\":\"Source of Truth vs evaluation ground truth\",\"level\":3},\"tunes\":{}},{\"id\":\"p-vs-ground-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"Evaluation ground truth is the reference answer, label or outcome against which a system is scored. It can be derived from authoritative sources, expert adjudication or curated test data.\"},\"tunes\":{}},{\"id\":\"p-vs-ground-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"Ground truth is therefore an evaluation construct. A Source of Truth is an application\u002Fdomain authority construct. They can overlap, but they are not interchangeable.\"},\"tunes\":{}},{\"id\":\"h-vs-quality\",\"type\":\"header\",\"data\":{\"text\":\"Source of Truth vs data quality\",\"level\":3},\"tunes\":{}},{\"id\":\"p-vs-quality-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"An authoritative source can still contain errors. Authority says which source officially governs the fact; data quality asks whether that source is accurate, complete, timely, consistent and fit for purpose.\"},\"tunes\":{}},{\"id\":\"p-vs-quality-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"When an authoritative system is known to be wrong, architecture should record the defect, correction process or exception instead of silently substituting an unofficial source and hiding the discrepancy.\"},\"tunes\":{}},{\"id\":\"h-model\",\"type\":\"header\",\"data\":{\"text\":\"A practical Source-of-Truth architecture model\",\"level\":2},\"tunes\":{}},{\"id\":\"model-note\",\"type\":\"callout\",\"data\":{\"variant\":\"note\",\"title\":\"Proposed architecture model\",\"body\":\"The model below is a practical synthesis for AI systems. It is not a W3C, NIST or ISO standard. It separates authority, provenance, retrieval and model-facing context because those responsibilities are often incorrectly collapsed.\"},\"tunes\":{}},{\"id\":\"sot-flow\",\"type\":\"processFlow\",\"data\":{\"title\":\"Authority-aware AI answer path\",\"orientation\":\"auto\",\"steps\":[{\"label\":\"1. Define the claim type\",\"description\":\"Identify what the user is actually asking: current state, policy, historical fact, technical specification, user preference, calculation or interpretation.\"},{\"label\":\"2. Resolve authority\",\"description\":\"Determine which source or authority class is permitted to define that type of claim for the required scope, version and time.\"},{\"label\":\"3. Acquire evidence\",\"description\":\"Read or retrieve the authoritative source and any necessary supporting or conflicting evidence.\"},{\"label\":\"4. Preserve provenance\",\"description\":\"Carry source identity, version, timestamp, locator, transformation history and responsibility metadata.\"},{\"label\":\"5. Build model context\",\"description\":\"Supply the relevant evidence to the model without discarding authority and applicability metadata.\"},{\"label\":\"6. Generate or compute\",\"description\":\"The model may summarize, compare, reason or transform the evidence, but it does not inherit source authority merely by processing it.\"},{\"label\":\"7. Validate the claim\",\"description\":\"Check that the answer is supported by the right source and remains inside its scope and validity boundary.\"},{\"label\":\"8. Preserve contradictions\",\"description\":\"If authoritative or relevant sources disagree, expose the conflict rather than fabricating false certainty.\"}]},\"tunes\":{}},{\"id\":\"h-registry\",\"type\":\"header\",\"data\":{\"text\":\"Authority should be explicit, not inferred from similarity\",\"level\":2},\"tunes\":{}},{\"id\":\"p-registry-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"One robust implementation pattern is an authority registry or equivalent policy layer that maps claim classes to authoritative source classes. The implementation can be code, metadata, configuration or domain rules; the important property is that authority is deliberate.\"},\"tunes\":{}},{\"id\":\"registry-table\",\"type\":\"table\",\"data\":{\"withHeadings\":true,\"stretched\":false,\"content\":[[\"Claim class\",\"Authority rule\",\"Fallback behavior\"],[\"Current account state\",\"Read live account service \u002F system of record\",\"If unavailable, report that current state cannot be verified.\"],[\"Product documentation\",\"Current approved documentation version\",\"Older version may be shown only with version warning.\"],[\"Implemented software behavior\",\"Relevant deployed release \u002F source artifact\",\"Documentation alone cannot prove deployed behavior.\"],[\"Internal policy\",\"Approved policy repository and active version\",\"Drafts are supporting material, not current authority.\"],[\"External technical standard\",\"Official standards body publication for relevant version\",\"Secondary explanations can clarify but not override the specification.\"],[\"Research claim\",\"Evidence policy appropriate to the domain\",\"Preserve conflicting evidence and confidence rather than force one source.\"]]},\"tunes\":{}},{\"id\":\"h-retrieval\",\"type\":\"header\",\"data\":{\"text\":\"Retrieval should use authority as a ranking constraint\",\"level\":2},\"tunes\":{}},{\"id\":\"p-retrieval-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"Semantic relevance answers “Which candidate looks related to this query?” Authority answers “Which candidate is allowed to establish this fact?” A production retrieval system often needs both.\"},\"tunes\":{}},{\"id\":\"p-retrieval-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"A useful sequence is to first constrain the candidate space by identity, tenant, source class, status, version or date, then rank relevant evidence inside the permitted space. If relevance is calculated before critical authorization or authority filters, the pipeline can return a convincing but invalid result.\"},\"tunes\":{}},{\"id\":\"relevance-authority\",\"type\":\"callout\",\"data\":{\"variant\":\"warning\",\"title\":\"Highest similarity is not highest authority\",\"body\":\"A search engine can rank a stale summary above the current primary source. A language model can prefer the better-written explanation over the official record. Neither behavior changes which source owns the fact.\"},\"tunes\":{}},{\"id\":\"h-freshness\",\"type\":\"header\",\"data\":{\"text\":\"Freshness is part of authority\",\"level\":2},\"tunes\":{}},{\"id\":\"p-freshness-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"Many Source-of-Truth failures are actually time failures. The correct source was known, but the system used an old snapshot, stale embedding, cached API response or superseded document.\"},\"tunes\":{}},{\"id\":\"p-freshness-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"An authority rule should therefore include invalidation or refresh semantics where the fact can change. “CRM is authoritative” is incomplete when the application reads a week-old replicated export.\"},\"tunes\":{}},{\"id\":\"h-derivation\",\"type\":\"header\",\"data\":{\"text\":\"Derived values need a trace back to authoritative inputs\",\"level\":2},\"tunes\":{}},{\"id\":\"p-derivation-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"Some important facts are not stored directly. They are computed from authoritative inputs: a risk score, account total, eligibility status or aggregate metric.\"},\"tunes\":{}},{\"id\":\"p-derivation-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"For derived values, Source-of-Truth architecture should preserve the input authorities, transformation or calculation version and execution time. W3C PROV's distinction between entities, activities and derivations is useful here because it models how one entity was produced from others.\"},\"tunes\":{}},{\"id\":\"derived-rule\",\"type\":\"callout\",\"data\":{\"variant\":\"success\",\"title\":\"Derived truth is conditional\",\"body\":\"A calculated value is only as current and authoritative as its inputs, transformation and validity conditions. Store the lineage needed to recompute or audit it.\"},\"tunes\":{}},{\"id\":\"h-conflict\",\"type\":\"header\",\"data\":{\"text\":\"What happens when authoritative sources disagree?\",\"level\":2},\"tunes\":{}},{\"id\":\"p-conflict-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"Conflicts are not edge cases in serious knowledge systems. A signed contract may disagree with a CRM field. Production behavior may disagree with documentation. Two primary historical sources may contradict each other. A current policy can conflict with an outdated local copy.\"},\"tunes\":{}},{\"id\":\"p-conflict-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"The system needs a resolution policy appropriate to the domain. Sometimes one authority clearly outranks the other. Sometimes the newer version supersedes the old one. Sometimes an expert or business owner must adjudicate. And sometimes the correct result is simply: the evidence is unresolved.\"},\"tunes\":{}},{\"id\":\"conflict-table\",\"type\":\"table\",\"data\":{\"withHeadings\":true,\"stretched\":false,\"content\":[[\"Conflict type\",\"Typical handling\"],[\"Current vs superseded version\",\"Use current version for present state; retain older version as historical evidence.\"],[\"System of record vs stale replica\",\"Use system of record; flag replication freshness issue.\"],[\"Contract vs CRM transcription\",\"Executed contract governs contractual wording; CRM discrepancy becomes a correction task.\"],[\"Documentation vs deployed behavior\",\"Distinguish intended behavior from observed\u002Fdeployed behavior; do not silently merge them.\"],[\"Two credible primary sources\",\"Preserve both, assess provenance and scope, and represent unresolved disagreement if no governing authority exists.\"],[\"User memory vs current user setting\",\"Use current explicit setting; mark memory as superseded where appropriate.\"]]},\"tunes\":{}},{\"id\":\"h-web\",\"type\":\"header\",\"data\":{\"text\":\"Web search is discovery, not automatically evidence\",\"level\":2},\"tunes\":{}},{\"id\":\"p-web-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"Search engines are excellent discovery systems. Search snippets, result ranking and generated summaries are not automatically primary evidence.\"},\"tunes\":{}},{\"id\":\"p-web-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"For claims that require authority, the search result should lead to the original publication, official record, source document, dataset or other appropriate artifact. The result page helps locate the source; it does not inherit the source's authority.\"},\"tunes\":{}},{\"id\":\"h-ai\",\"type\":\"header\",\"data\":{\"text\":\"The language model should not decide authority by itself\",\"level\":2},\"tunes\":{}},{\"id\":\"p-ai-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"A model can help classify a question, extract claims or compare evidence, but authority should not depend only on the model's preference. Models optimize generation from context; they do not possess a guaranteed domain-specific registry of which database, document or organization owns each fact.\"},\"tunes\":{}},{\"id\":\"p-ai-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"This is why application architecture should encode critical authority rules deterministically where practical. The model may reason within the boundary, but the boundary itself should not be recreated from scratch for every prompt.\"},\"tunes\":{}},{\"id\":\"h-observability\",\"type\":\"header\",\"data\":{\"text\":\"Authority must survive the execution trace\",\"level\":2},\"tunes\":{}},{\"id\":\"p-observability-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"If a production answer is important enough to audit, the trace should make it possible to reconstruct which sources were consulted, which version was used, which passage or record supported the claim, which transformations occurred and whether conflicting evidence was available.\"},\"tunes\":{}},{\"id\":\"p-observability-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"This aligns with the broader provenance principle in W3C PROV and with NIST AI RMF Playbook guidance to document sources, origins, transformations, dependencies, constraints and metadata.\"},\"tunes\":{}},{\"id\":\"h-implementation\",\"type\":\"header\",\"data\":{\"text\":\"Original implementation evidence: Source of Truth Research Engine\",\"level\":2},\"tunes\":{}},{\"id\":\"implementation-note\",\"type\":\"callout\",\"data\":{\"variant\":\"note\",\"title\":\"Original implementation evidence\",\"body\":\"The Source of Truth Research Engine is my own implementation evidence for an evidence-first architecture. It demonstrates one concrete way to separate discovery, source acquisition, provenance, claims, evidence classes, contradictions and conclusions. It is an implementation pattern, not a universal industry standard.\"},\"tunes\":{}},{\"id\":\"p-impl-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"The engine is designed around a traceable pipeline rather than direct AI summarization: research task → search → original source or digital artifact → local snapshot → SHA-256 → source ID → claim → evidence class → relation or contradiction → interpretation → conclusion.\"},\"tunes\":{}},{\"id\":\"p-impl-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"Its shared evidence core stores Sources, Artifacts, provenance, Claims, Relations, Contradictions, a Reference Model and an audit trail. Different research modes can share that core while applying different domain methodologies.\"},\"tunes\":{}},{\"id\":\"p-impl-3\",\"type\":\"paragraph\",\"data\":{\"text\":\"The architecture deliberately separates discovery from evidence. Search snippets are not treated as evidence, filenames are not treated as content, AI summaries are not treated as primary sources, and semantic similarity is only a discovery signal until a result is traced back to a concrete source and locator.\"},\"tunes\":{}},{\"id\":\"p-impl-4\",\"type\":\"paragraph\",\"data\":{\"text\":\"Original files are preserved and local bytes receive SHA-256 identifiers. Contradictions and rejected hypotheses are not silently deleted. New evidence is allowed to change the current reference model while the prior evidence path remains auditable.\"},\"tunes\":{}},{\"id\":\"implementation-table\",\"type\":\"table\",\"data\":{\"withHeadings\":true,\"stretched\":false,\"content\":[[\"Implemented rule\",\"Why it matters for Source-of-Truth architecture\"],[\"Search ≠ evidence\",\"Discovery ranking cannot silently become authority.\"],[\"Filename ≠ content\",\"Metadata clues cannot substitute for reading the actual artifact.\"],[\"Local snapshot + SHA-256\",\"Evidence can be tied to exact bytes rather than a mutable remote label.\"],[\"Source ID + exact locator\",\"Claims can be traced back to the concrete evidence location.\"],[\"Claim\u002Fevidence separation\",\"The assertion is not confused with the material supporting it.\"],[\"Contradictions preserved\",\"The system can represent unresolved disagreement rather than overwrite history.\"],[\"Semantic similarity is discovery only\",\"Retrieval relevance is explicitly separated from evidentiary authority.\"],[\"New evidence may update the model\",\"Source-of-Truth state is versioned and revisable rather than treated as immutable dogma.\"]]},\"tunes\":{}},{\"id\":\"h-document-engine\",\"type\":\"header\",\"data\":{\"text\":\"Aaasaasa Document & Knowledge Engine: applying the same boundary to enterprise documents\",\"level\":2},\"tunes\":{}},{\"id\":\"p-doc-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"The Aaasaasa Document & Knowledge Engine concept extends the same design principle to enterprise documentation: users should be able to search documents, ask source-grounded questions and review collections against explicit criteria while preserving the distinction between what a document states and what the system infers.\"},\"tunes\":{}},{\"id\":\"p-doc-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"The important architecture rule is that a universal retrieval core does not make every collection equally authoritative. Contract documents, maintenance records, financial documents and research material need different authority, validation and coverage rules even when they share ingestion and search infrastructure.\"},\"tunes\":{}},{\"id\":\"h-failures\",\"type\":\"header\",\"data\":{\"text\":\"Common Source-of-Truth failure modes\",\"level\":2},\"tunes\":{}},{\"id\":\"failures-table\",\"type\":\"table\",\"data\":{\"withHeadings\":true,\"stretched\":false,\"content\":[[\"Failure mode\",\"What goes wrong\"],[\"The model is treated as the source of truth\",\"Parametric knowledge can be stale, incomplete, unverifiable or outside the application's authoritative scope.\"],[\"Top retrieval result wins automatically\",\"Similarity is mistaken for authority.\"],[\"Vector database becomes authoritative\",\"Derived index records lose the identity and version of the original source.\"],[\"Everything is copied into one knowledge base\",\"Copies obscure ownership, freshness and correction paths.\"],[\"Memory is reused as current state\",\"Old observations silently override the current system of record.\"],[\"No version metadata\",\"The right document is used for the wrong time period.\"],[\"No source locator\",\"A citation exists but the supporting passage or record cannot be verified.\"],[\"Conflicts are overwritten\",\"The system appears consistent by destroying evidence of disagreement.\"],[\"Generated summaries replace originals\",\"A lossy transformation becomes the apparent authority.\"],[\"Authority is global instead of claim-specific\",\"One source is trusted beyond the domain or fact class it actually owns.\"],[\"Web snippet is treated as evidence\",\"Discovery metadata replaces the original publication.\"],[\"Authoritative data is wrong but exceptions are hidden\",\"Operational defects become invisible and cannot be corrected transparently.\"]]},\"tunes\":{}},{\"id\":\"h-decision\",\"type\":\"header\",\"data\":{\"text\":\"A practical Source-of-Truth decision framework\",\"level\":2},\"tunes\":{}},{\"id\":\"decision-flow\",\"type\":\"processFlow\",\"data\":{\"title\":\"How to decide what should define a claim\",\"orientation\":\"auto\",\"steps\":[{\"label\":\"1. State the claim precisely\",\"description\":\"Separate current state, historical state, policy, interpretation, prediction and derived calculation.\"},{\"label\":\"2. Identify the authority owner\",\"description\":\"Determine the system, document, institution, person or evidence class responsible for that claim type.\"},{\"label\":\"3. Define scope\",\"description\":\"Specify tenant, jurisdiction, product, environment, user, document set or other applicability boundary.\"},{\"label\":\"4. Define time and version\",\"description\":\"Determine whether the claim requires current state, a historical snapshot or a specific standard\u002Frelease version.\"},{\"label\":\"5. Preserve provenance\",\"description\":\"Record source identity, origin, locator, transformations and responsible agents or processes.\"},{\"label\":\"6. Define retrieval\u002Faccess path\",\"description\":\"Make sure the application can actually obtain the authoritative information under the correct identity and permissions.\"},{\"label\":\"7. Define conflict policy\",\"description\":\"Decide precedence, supersession, adjudication or explicit unresolved-state behavior.\"},{\"label\":\"8. Define invalidation\",\"description\":\"Specify when cached, indexed, remembered or derived representations must be refreshed.\"},{\"label\":\"9. Validate the answer path\",\"description\":\"Verify that important generated claims can be traced to the intended authority, not merely to a plausible source.\"}]},\"tunes\":{}},{\"id\":\"h-checklist\",\"type\":\"header\",\"data\":{\"text\":\"Source-of-Truth architecture checklist\",\"level\":2},\"tunes\":{}},{\"id\":\"checklist-table\",\"type\":\"table\",\"data\":{\"withHeadings\":true,\"stretched\":false,\"content\":[[\"Question\",\"Expected answer\"],[\"What exact fact or state is being established?\",\"A claim precise enough to assign authority.\"],[\"Who or what owns that fact?\",\"Named authoritative system, source class or adjudication rule.\"],[\"Is the authority current for this scope?\",\"Tenant, jurisdiction, environment, user or domain boundary is explicit.\"],[\"Is the version\u002Ftime correct?\",\"Current, historical or version-specific applicability is known.\"],[\"Can the source be verified?\",\"Stable identifier, locator or record reference exists.\"],[\"Is provenance preserved?\",\"Origin, transformation and responsibility metadata survive ingestion and retrieval.\"],[\"Can retrieval return non-authoritative material?\",\"If yes, filters or validation distinguish relevance from authority.\"],[\"Can the source change?\",\"Refresh, invalidation or supersession rules exist.\"],[\"Can sources disagree?\",\"Conflict and adjudication behavior is explicit.\"],[\"Can memory become stale?\",\"Volatile state is re-read from the current authority before consequential use.\"],[\"Can a derived answer be reproduced?\",\"Inputs, transformation version and execution conditions are traceable.\"],[\"Can an auditor reconstruct the answer?\",\"Execution evidence preserves the source path for important claims.\"]]},\"tunes\":{}},{\"id\":\"h-misconceptions\",\"type\":\"header\",\"data\":{\"text\":\"Common misconceptions\",\"level\":2},\"tunes\":{}},{\"id\":\"misconceptions-table\",\"type\":\"table\",\"data\":{\"withHeadings\":true,\"stretched\":false,\"content\":[[\"Misconception\",\"Correction\"],[\"“Source of Truth means one database.”\",\"One database can be authoritative for one domain; complex systems usually have multiple fact-specific authorities.\"],[\"“The newest document is automatically authoritative.”\",\"Recency helps only when the newer artifact is approved and actually supersedes the older one.\"],[\"“RAG solves truth.”\",\"RAG solves retrieval. Authority, provenance, evidence quality and validity remain separate problems.\"],[\"“A citation proves the answer.”\",\"The cited source must actually support the claim, have the right authority and apply to the current scope.\"],[\"“Provenance tells us what is true.”\",\"Provenance tells us origin and derivation; authority and correctness still require domain rules and evaluation.\"],[\"“The system of record is always correct.”\",\"It is authoritative for the operational record, but data-quality defects can still exist and require visible correction.\"],[\"“If several sources agree, the claim is authoritative.”\",\"Agreement increases evidence but does not necessarily establish ownership or applicability.\"],[\"“AI memory can replace repeated reads.”\",\"Only for information whose staleness risk is acceptable; volatile or consequential state should be refreshed from authority.\"]]},\"tunes\":{}},{\"id\":\"h-edge\",\"type\":\"header\",\"data\":{\"text\":\"Edge cases\",\"level\":2},\"tunes\":{}},{\"id\":\"p-edge-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"Some questions are interpretive rather than factual. “Which architecture is best?” has no single Source of Truth. The system can retrieve authoritative constraints and evidence, but the final judgment is an inference that should expose assumptions and trade-offs.\"},\"tunes\":{}},{\"id\":\"p-edge-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"Some domains use distributed authority. A scientific conclusion can depend on multiple studies, datasets and replications. A historical conclusion can depend on conflicting primary and secondary evidence. The architecture should represent evidence structure rather than invent a central database that supposedly owns truth.\"},\"tunes\":{}},{\"id\":\"p-edge-3\",\"type\":\"paragraph\",\"data\":{\"text\":\"A user can also be the authority for subjective personal information: preferences, goals, chosen settings or explicit instructions. Even then, newer explicit input can supersede older memory.\"},\"tunes\":{}},{\"id\":\"p-edge-4\",\"type\":\"paragraph\",\"data\":{\"text\":\"External events can invalidate previously authoritative data. A price feed, inventory system or security policy may have been correct when captured but no longer valid. Snapshot provenance preserves what was true then; it does not make the snapshot current forever.\"},\"tunes\":{}},{\"id\":\"h-limit\",\"type\":\"header\",\"data\":{\"text\":\"Limitations\",\"level\":2},\"tunes\":{}},{\"id\":\"p-limit-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"Source-of-Truth architecture cannot guarantee that an authoritative source is factually correct. It provides accountability, provenance and deterministic ownership boundaries; data-quality and domain-verification processes remain necessary.\"},\"tunes\":{}},{\"id\":\"p-limit-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"Authority can also be contested. Different institutions can legitimately claim authority in different jurisdictions or methodologies. In those situations the system should expose the authority model and disagreement rather than hide it behind a universal “truth score.”\"},\"tunes\":{}},{\"id\":\"p-limit-3\",\"type\":\"paragraph\",\"data\":{\"text\":\"Finally, authority rules require maintenance. Systems, owners, policies, versions and regulations change. A stale authority registry can be as dangerous as no registry at all.\"},\"tunes\":{}},{\"id\":\"h-change\",\"type\":\"header\",\"data\":{\"text\":\"What would change this answer?\",\"level\":2},\"tunes\":{}},{\"id\":\"p-change-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"The specific authority mapping changes with the domain. Banking, healthcare, software delivery, scientific research and historical analysis have different systems of record, evidentiary rules and regulatory obligations.\"},\"tunes\":{}},{\"id\":\"p-change-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"The implementation also changes with architecture. A small application can encode authority directly in service calls. A larger platform may need registries, source metadata, policy engines, lineage systems or data contracts. The core principle remains the same: do not let retrieval order or model preference silently decide what counts as authoritative.\"},\"tunes\":{}},{\"id\":\"h-related\",\"type\":\"header\",\"data\":{\"text\":\"Related canonical knowledge\",\"level\":2},\"tunes\":{}},{\"id\":\"p-related-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"Source-of-Truth architecture is a prerequisite for later retrieval and governance concepts because retrieval quality alone cannot determine whether evidence is allowed to define the answer.\"},\"tunes\":{}},{\"id\":\"ref-rag\",\"type\":\"referralArticle\",\"data\":{\"url\":\"https:\u002F\u002Fstajic.de\u002Fblog\u002Fwhat-is-rag-the-simplest-explanation-of-how-it-works\",\"title\":\"What Is RAG? The Simplest Explanation of How It Works\",\"excerpt\":\"RAG retrieves external knowledge for a model. This foundation explains why retrieval and authoritative truth are separate responsibilities.\",\"ctaLabel\":\"Read the RAG foundation\"},\"tunes\":{}},{\"id\":\"p-related-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"Memory is another adjacent concept. A reliable agent separates remembered information from current authoritative application state.\"},\"tunes\":{}},{\"id\":\"ref-memory\",\"type\":\"referralArticle\",\"data\":{\"url\":\"https:\u002F\u002Fstajic.de\u002Fblog\u002Fai-agent-memory-is-not-rag-how-to-separate-memory-retrieval-state-and-context\",\"title\":\"AI Agent Memory Is Not RAG: How to Separate Memory, Retrieval, State and Context\",\"excerpt\":\"A four-layer architecture separating persistent memory, authoritative state, retrieval and the context actually supplied to the model.\",\"ctaLabel\":\"Read the memory architecture article\"},\"tunes\":{}},{\"id\":\"p-related-3\",\"type\":\"paragraph\",\"data\":{\"text\":\"Authority also connects directly to answer validity. Even an authoritative source supports only claims inside its version, date, scope and evidence boundary.\"},\"tunes\":{}},{\"id\":\"ref-avb\",\"type\":\"referralArticle\",\"data\":{\"url\":\"https:\u002F\u002Fstajic.de\u002Fblog\u002Fthe-answer-validity-boundary-the-missing-layer-between-relevance-and-reliable-ai-answers\",\"title\":\"The Answer Validity Boundary: The Missing Layer Between Relevance and Reliable AI Answers\",\"excerpt\":\"A framework for making explicit where an AI answer remains supported by evidence and which changed conditions invalidate that support.\",\"ctaLabel\":\"Read the Answer Validity Boundary\"},\"tunes\":{}},{\"id\":\"h-faq\",\"type\":\"header\",\"data\":{\"text\":\"Frequently asked questions\",\"level\":2},\"tunes\":{}},{\"id\":\"faq\",\"type\":\"faq\",\"data\":{\"title\":\"Source of Truth in AI systems\",\"items\":[{\"id\":\"faq1\",\"question\":\"What is a Source of Truth in an AI system?\",\"answer\":\"It is the authoritative source or authority rule that determines which source is allowed to establish a specific fact, state or rule for a defined scope, version and time.\"},{\"id\":\"faq2\",\"question\":\"Is the language model a Source of Truth?\",\"answer\":\"Normally no. A language model can generate, summarize and reason, but its parametric knowledge is not automatically authoritative for current application state, company policy, a specific document version or a regulated domain fact.\"},{\"id\":\"faq3\",\"question\":\"Is a vector database the Source of Truth for RAG?\",\"answer\":\"Not automatically. A vector database is usually an index or retrieval store. It should preserve references to the authoritative original source and version rather than silently replacing them.\"},{\"id\":\"faq4\",\"question\":\"What is the difference between provenance and Source of Truth?\",\"answer\":\"Provenance describes where data came from, how it was produced or transformed and who or what was involved. Source-of-Truth rules determine whether that source has authority for the specific claim.\"},{\"id\":\"faq5\",\"question\":\"Can an AI system have multiple Sources of Truth?\",\"answer\":\"Yes. In complex systems this is normal because different facts belong to different authoritative systems or source classes.\"},{\"id\":\"faq6\",\"question\":\"What happens when two authoritative sources disagree?\",\"answer\":\"The system needs a domain-specific conflict rule: precedence, version supersession, expert adjudication or an explicit unresolved state. It should not silently choose the source the model prefers.\"},{\"id\":\"faq7\",\"question\":\"Does RAG guarantee that an AI answer uses the Source of Truth?\",\"answer\":\"No. RAG retrieves candidates. Authority-aware metadata, filters, source policy and validation are needed to ensure that important claims use the correct source.\"},{\"id\":\"faq8\",\"question\":\"Can the Source of Truth be wrong?\",\"answer\":\"Yes. Authority and correctness are different properties. An authoritative system can contain a data-quality defect, which should be corrected transparently rather than hidden by substituting an unofficial source.\"}]},\"tunes\":{}},{\"id\":\"h-glossary\",\"type\":\"header\",\"data\":{\"text\":\"Glossary\",\"level\":2},\"tunes\":{}},{\"id\":\"glossary\",\"type\":\"glossary\",\"data\":{\"title\":\"Key Source-of-Truth terms\",\"entries\":[{\"term\":\"Source of Truth\",\"definition\":\"The authoritative source or rule permitted to establish a particular fact, state or rule for a defined scope, version and time.\",\"anchor\":\"source-of-truth\"},{\"term\":\"System of record\",\"definition\":\"The authoritative operational system responsible for a defined class of records or current business state.\",\"anchor\":\"system-of-record\"},{\"term\":\"Provenance\",\"definition\":\"Information describing the origin, derivation, transformations, responsible agents and history of data or another entity.\",\"anchor\":\"provenance\"},{\"term\":\"Evidence\",\"definition\":\"Information or an artifact that supports, contradicts or constrains a claim.\",\"anchor\":\"evidence\"},{\"term\":\"Authority\",\"definition\":\"The application or domain rule determining which source is entitled to define a specific claim.\",\"anchor\":\"authority\"},{\"term\":\"Freshness\",\"definition\":\"Whether a representation remains current enough for the claim or operation in which it is used.\",\"anchor\":\"freshness\"},{\"term\":\"Supersession\",\"definition\":\"The explicit replacement of an older authoritative version by a newer one while preserving historical traceability.\",\"anchor\":\"supersession\"},{\"term\":\"Ground truth\",\"definition\":\"A reference answer, label or outcome used to evaluate a system; it is an evaluation construct rather than automatically the application's Source of Truth.\",\"anchor\":\"ground-truth\"},{\"term\":\"Lineage\",\"definition\":\"The trace of how data or derived values flow and transform across sources and processing steps.\",\"anchor\":\"lineage\"},{\"term\":\"Validity boundary\",\"definition\":\"The conditions of scope, time, version, evidence and assumptions inside which a claim remains supported.\",\"anchor\":\"validity-boundary\"}]},\"tunes\":{}},{\"id\":\"h-conclusion\",\"type\":\"header\",\"data\":{\"text\":\"Conclusion\",\"level\":2},\"tunes\":{}},{\"id\":\"p-conclusion-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"Reliable AI does not come from giving the model more information. It comes from knowing which information is allowed to define the claim, preserving where that information came from, retrieving the correct version and keeping the final answer inside the source's scope.\"},\"tunes\":{}},{\"id\":\"p-conclusion-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"That is why Source of Truth, provenance, retrieval, memory and context must remain separate concepts. The Source of Truth defines authority. Provenance explains origin. Retrieval finds candidates. Memory preserves selected past information. Context is what the model sees. Generation turns those inputs into an output.\"},\"tunes\":{}},{\"id\":\"p-conclusion-3\",\"type\":\"paragraph\",\"data\":{\"text\":\"When those layers remain explicit, an AI system can do more than sound plausible: important claims can be traced back to the source that actually had the right to establish them.\"},\"tunes\":{}},{\"id\":\"h-sources\",\"type\":\"header\",\"data\":{\"text\":\"Primary sources and implementation evidence\",\"level\":2},\"tunes\":{}},{\"id\":\"p-sources-note\",\"type\":\"paragraph\",\"data\":{\"text\":\"The external sources below support provenance and AI risk-management claims. The Source of Truth Research Engine section is original implementation evidence and is explicitly presented as one implementation pattern rather than a universal standard.\"},\"tunes\":{}},{\"id\":\"src-w3c-prov\",\"type\":\"linkTool\",\"data\":{\"link\":\"https:\u002F\u002Fwww.w3.org\u002FTR\u002Fprov-dm\u002F\",\"meta\":{\"image\":{\"url\":\"\"},\"title\":\"W3C PROV-DM — The PROV Data Model\",\"description\":\"W3C Recommendation defining a domain-agnostic provenance model around entities, activities, agents, derivations and responsibility.\"}},\"tunes\":{}},{\"id\":\"src-w3c-publications\",\"type\":\"linkTool\",\"data\":{\"link\":\"https:\u002F\u002Fwww.w3.org\u002Fgroups\u002Fwg\u002Fprov\u002Fpublications\u002F\",\"meta\":{\"image\":{\"url\":\"\"},\"title\":\"W3C Provenance Working Group — Publications\",\"description\":\"Official index of W3C PROV Recommendations and related specifications for provenance interchange and constraints.\"}},\"tunes\":{}},{\"id\":\"src-nist-rmf\",\"type\":\"linkTool\",\"data\":{\"link\":\"https:\u002F\u002Fwww.nist.gov\u002Fitl\u002Fai-risk-management-framework\",\"meta\":{\"image\":{\"url\":\"\"},\"title\":\"NIST AI Risk Management Framework\",\"description\":\"NIST's voluntary framework for incorporating trustworthiness and risk-management considerations across the AI lifecycle; AI RMF 1.0 is currently being revised.\"}},\"tunes\":{}},{\"id\":\"src-nist-playbook\",\"type\":\"linkTool\",\"data\":{\"link\":\"https:\u002F\u002Fairc.nist.gov\u002Fairmf-resources\u002Fplaybook\u002F\",\"meta\":{\"image\":{\"url\":\"\"},\"title\":\"NIST AI RMF Playbook\",\"description\":\"Operational guidance aligned to the AI RMF, including documentation practices for data provenance, sources, origins, transformations, dependencies, constraints and metadata.\"}},\"tunes\":{}},{\"id\":\"src-nist-measure\",\"type\":\"linkTool\",\"data\":{\"link\":\"https:\u002F\u002Fairc.nist.gov\u002Fairmf-resources\u002Fplaybook\u002Fmeasure\u002F\",\"meta\":{\"image\":{\"url\":\"\"},\"title\":\"NIST AI RMF Playbook — Measure\",\"description\":\"Guidance on documenting measurement, data provenance and contextual interpretation of AI system outputs.\"}},\"tunes\":{}},{\"id\":\"src-nist-genai\",\"type\":\"linkTool\",\"data\":{\"link\":\"https:\u002F\u002Fwww.nist.gov\u002Fpublications\u002Fartificial-intelligence-risk-management-framework-generative-artificial-intelligence\",\"meta\":{\"image\":{\"url\":\"\"},\"title\":\"NIST AI 600-1 — Generative AI Profile\",\"description\":\"NIST generative-AI profile, including provenance and information-integrity considerations for generative AI systems.\"}},\"tunes\":{}}],\"version\":\"2.31.6\"}",{"time":1337,"blocks":1338,"version":2243},1791479104597,[1339,1343,1348,1353,1358,1362,1366,1370,1374,1378,1383,1387,1391,1395,1399,1423,1427,1431,1435,1439,1475,1479,1483,1487,1491,1495,1499,1503,1507,1511,1515,1519,1523,1527,1531,1535,1539,1543,1547,1551,1555,1559,1563,1567,1571,1575,1579,1583,1587,1591,1595,1600,1629,1633,1637,1669,1673,1677,1681,1686,1690,1694,1698,1702,1706,1710,1715,1719,1723,1727,1752,1756,1760,1764,1768,1772,1776,1780,1784,1788,1792,1797,1801,1805,1809,1813,1844,1848,1852,1856,1860,1903,1907,1939,1943,1985,1989,2020,2024,2028,2032,2036,2040,2044,2048,2052,2056,2060,2064,2068,2072,2076,2083,2087,2094,2098,2105,2109,2138,2142,2177,2181,2185,2189,2193,2197,2201,2208,2215,2222,2229,2236],{"id":215,"data":1340,"type":218,"tunes":1342},{"text":1341},"A Source of Truth in an AI system is the authoritative source allowed to define whether a specific fact, state or rule should be treated as true for a particular scope, version and time. It is not automatically the language model, the vector database, the top-ranked retrieved document, agent memory or the latest message in context. Reliable AI architecture must preserve which source has authority for which claim, then keep provenance, retrieval and validation connected to that authority.",{},{"id":221,"data":1344,"type":226,"tunes":1347},{"body":1345,"title":1346,"variant":225},"\u003Cstrong>The Source of Truth is an authority rule, not an AI component.\u003C\u002Fstrong> An AI system can search many sources, but only some sources are authoritative for a given question. The correct source may be an application database for current account state, an approved policy document for a rule, source code for implemented behavior, an official specification for a protocol, or a primary research artifact for an observed claim. Good architecture makes that authority explicit instead of asking the model to infer it from relevance.","Direct answer",{},{"id":229,"data":1349,"type":226,"tunes":1352},{"body":1350,"title":1351,"variant":233},"“Source of Truth” is widely used in software, data and organizational practice, but it is not a standardized AI component with one universal formal definition. This article uses it as an architecture concept: \u003Cstrong>the source or authority mapping that determines which evidence is allowed to establish a particular fact or state\u003C\u002Fstrong>. Provenance standards such as W3C PROV describe origin and derivation; they do not automatically decide which source your application should treat as authoritative.","Terminology boundary",{},{"id":236,"data":1354,"type":226,"tunes":1357},{"body":1355,"title":1356,"variant":233},"The core distinction between authority, provenance, retrieval and model context is stable. NIST AI RMF 1.0 is currently being revised, but the current NIST Playbook still explicitly recommends documenting data provenance, including sources, origins, transformations, dependencies, constraints and metadata. W3C PROV remains a W3C Recommendation for interoperable provenance descriptions.","Current-source note — 8 October 2026",{},{"id":242,"data":1359,"type":247,"tunes":1361},{"title":1360,"maxLevel":245,"minLevel":246},"Contents",{},{"id":250,"data":1363,"type":42,"tunes":1365},{"text":1364,"level":246},"What “Source of Truth” really means",{},{"id":255,"data":1367,"type":218,"tunes":1369},{"text":1368},"The phrase is often misunderstood as “the one database that contains everything.” That can be true in a narrow system, but it is usually too simplistic for AI. A real AI application can combine operational databases, documents, APIs, vector indexes, user input, model memory, external web sources and generated summaries.",{},{"id":260,"data":1371,"type":218,"tunes":1373},{"text":1372},"Those sources do not have equal authority. A customer-support manual may define policy but not a customer's current balance. A CRM may define the current account owner but not the legal meaning of a regulation. A source-code repository may define implemented behavior while a product specification defines intended behavior. The architecture must therefore answer a more precise question: which source is authoritative for this specific claim?",{},{"id":265,"data":1375,"type":218,"tunes":1377},{"text":1376},"This makes Source of Truth a relationship between a claim and an authority, not merely a property of a storage technology.",{},{"id":270,"data":1379,"type":226,"tunes":1382},{"body":1380,"title":1381,"variant":274},"\u003Cstrong>One system can have many Sources of Truth because authority is fact-specific.\u003C\u002Fstrong> The important architectural property is not that one database wins globally, but that each important fact or state has an explicit authoritative owner.","Core principle",{},{"id":277,"data":1384,"type":42,"tunes":1386},{"text":1385,"level":246},"The simplest example",{},{"id":282,"data":1388,"type":218,"tunes":1390},{"text":1389},"A user asks an AI assistant: “What is my current subscription plan?” The assistant has three possible inputs: last month's support transcript, an indexed help-center document describing plan types, and the live billing database.",{},{"id":287,"data":1392,"type":218,"tunes":1394},{"text":1393},"The support transcript may mention that the user had a Pro plan. The help-center document explains what Pro means. But the live billing record is the authoritative source for the user's current subscription state.",{},{"id":292,"data":1396,"type":218,"tunes":1398},{"text":1397},"A semantic search engine could rank the support transcript above the billing record because it contains language closer to the question. That ranking would still not make the transcript authoritative. Relevance and authority are different dimensions.",{},{"id":297,"data":1400,"type":329,"tunes":1422},{"rows":1401,"title":1414,"layout":318,"columns":1415},[1402,1405,1408,1411],{"id":301,"label":1403,"values":1404},"Billing database",[304,304,304],{"id":306,"label":1406,"values":1407},"Help-center documentation",[304,304,304],{"id":310,"label":1409,"values":1410},"Old support transcript",[304,304,304],{"id":314,"label":1412,"values":1413},"Model memory",[304,304,304],"The same question can involve different source roles",[1416,1418,1420],{"id":321,"label":1417},"Source",{"id":324,"label":1419},"Role",{"id":327,"label":1421},"Authority for current plan?",{},{"id":332,"data":1424,"type":42,"tunes":1426},{"text":1425,"level":246},"Where the simple example stops",{},{"id":337,"data":1428,"type":218,"tunes":1430},{"text":1429},"Not every domain has one unquestioned authority. Historical research can contain conflicting primary sources. Scientific claims can evolve as new studies appear. Legal interpretation can depend on jurisdiction, date and court authority. Product behavior can differ between documentation and deployed code.",{},{"id":342,"data":1432,"type":218,"tunes":1434},{"text":1433},"In these cases the correct architecture is not to invent a single winner. It is to preserve the competing sources, their provenance, their authority class, their applicable scope and the unresolved contradiction. A reliable Source-of-Truth system must be able to represent uncertainty and disagreement.",{},{"id":347,"data":1436,"type":42,"tunes":1438},{"text":1437,"level":246},"Authority is scoped by claim, version and time",{},{"id":352,"data":1440,"type":318,"tunes":1474},{"content":1441,"stretched":43,"withHeadings":14},[1442,1446,1450,1454,1458,1462,1466,1470],[1443,1444,1445],"Question","Possible authoritative source","Why scope matters",[1447,1448,1449],"What is the user's current account balance?","Ledger \u002F accounting system of record","Historical exports may be accurate for an earlier time but not current state.",[1451,1452,1453],"What does company policy currently allow?","Approved current policy version","An older policy may remain valid evidence of past rules but not present rules.",[1455,1456,1457],"What code is actually deployed?","Deployment artifact \u002F commit \u002F release record","Main branch may differ from production.",[1459,1460,1461],"What did a contract state when signed?","Executed contract version","A draft or later template is not authoritative for the signed agreement.",[1463,1464,1465],"What does a technical protocol specify?","Current official specification for the relevant version","A blog explanation may be useful but is secondary evidence.",[1467,1468,1469],"What happened in a historical event?","Relevant primary evidence plus explicit source criticism","There may be no single authority; conflicting evidence must remain visible.",[1471,1472,1473],"What does a user prefer?","Current explicit user setting or confirmed preference","Old conversation memory may be stale or superseded.",{},{"id":389,"data":1476,"type":218,"tunes":1478},{"text":1477},"The word “truth” can therefore be misleading unless its boundary is stated. In architecture, the Source of Truth is usually better understood as the source authorized to determine a specific proposition under defined conditions.",{},{"id":394,"data":1480,"type":42,"tunes":1482},{"text":1481,"level":246},"What a Source of Truth is not",{},{"id":399,"data":1484,"type":42,"tunes":1486},{"text":1485,"level":245},"Source of Truth vs system of record",{},{"id":404,"data":1488,"type":218,"tunes":1490},{"text":1489},"A system of record is typically the authoritative operational system for a class of records: for example, a billing ledger, HR master record or order database. It is one common implementation of Source-of-Truth authority.",{},{"id":409,"data":1492,"type":218,"tunes":1494},{"text":1493},"But Source of Truth is broader. A signed PDF contract, an official standard, a deployment artifact or a primary archival document may be authoritative without being a transactional system of record.",{},{"id":414,"data":1496,"type":42,"tunes":1498},{"text":1497,"level":245},"Source of Truth vs provenance",{},{"id":419,"data":1500,"type":218,"tunes":1502},{"text":1501},"Provenance answers questions such as: Where did this data come from? Who or what produced it? Which transformation created this derivative? Which prior entity was used? W3C PROV models entities, activities, agents and derivations so origin and responsibility can be represented.",{},{"id":424,"data":1504,"type":218,"tunes":1506},{"text":1505},"Provenance does not by itself establish authority. Knowing that a value came from a spreadsheet written by a specific employee helps evaluate it, but the application still needs a rule saying whether that spreadsheet is authoritative for the claim.",{},{"id":429,"data":1508,"type":42,"tunes":1510},{"text":1509,"level":245},"Source of Truth vs evidence",{},{"id":434,"data":1512,"type":218,"tunes":1514},{"text":1513},"Evidence supports or contradicts a claim. A Source of Truth defines which source has the authority to settle or strongly constrain that claim in the current application context.",{},{"id":439,"data":1516,"type":218,"tunes":1518},{"text":1517},"A source can be valuable evidence without being authoritative. Five customer emails may be evidence that users dislike a workflow, but they are not the system of record for current product configuration.",{},{"id":444,"data":1520,"type":42,"tunes":1522},{"text":1521,"level":245},"Source of Truth vs RAG",{},{"id":449,"data":1524,"type":218,"tunes":1526},{"text":1525},"RAG is a retrieval pattern. It finds information and supplies selected content to the model. RAG does not automatically know which source deserves authority.",{},{"id":454,"data":1528,"type":218,"tunes":1530},{"text":1529},"A RAG pipeline can retrieve a stale document, a secondary summary or a highly similar but non-authoritative source. Source authority must be encoded through corpus design, metadata, filters, ranking policy, validation or post-retrieval checks.",{},{"id":459,"data":1532,"type":42,"tunes":1534},{"text":1533,"level":245},"Source of Truth vs vector database",{},{"id":464,"data":1536,"type":218,"tunes":1538},{"text":1537},"A vector database stores or indexes representations used for semantic retrieval. It is an access layer, not automatically a truth layer.",{},{"id":469,"data":1540,"type":218,"tunes":1542},{"text":1541},"The same authoritative document may be chunked, embedded, copied and re-indexed many times. The vector record should retain a reference back to the authoritative source and version rather than becoming an untraceable new authority.",{},{"id":474,"data":1544,"type":42,"tunes":1546},{"text":1545,"level":245},"Source of Truth vs memory",{},{"id":479,"data":1548,"type":218,"tunes":1550},{"text":1549},"Agent or application memory stores information that may be useful later. Memory can preserve a prior decision, preference or observation, but it can become stale.",{},{"id":484,"data":1552,"type":218,"tunes":1554},{"text":1553},"For volatile or consequential state, a reliable agent should normally re-read the authoritative current source rather than assume that remembered state is still true.",{},{"id":489,"data":1556,"type":42,"tunes":1558},{"text":1557,"level":245},"Source of Truth vs context",{},{"id":494,"data":1560,"type":218,"tunes":1562},{"text":1561},"Context is what the model receives during the current inference. Authoritative information can be absent from context, while non-authoritative information can be present.",{},{"id":499,"data":1564,"type":218,"tunes":1566},{"text":1565},"Context construction therefore needs an authority-aware policy: retrieve or read the source that is allowed to define the claim, then preserve enough metadata for the model or validator to understand its scope.",{},{"id":504,"data":1568,"type":42,"tunes":1570},{"text":1569,"level":245},"Source of Truth vs evaluation ground truth",{},{"id":509,"data":1572,"type":218,"tunes":1574},{"text":1573},"Evaluation ground truth is the reference answer, label or outcome against which a system is scored. It can be derived from authoritative sources, expert adjudication or curated test data.",{},{"id":514,"data":1576,"type":218,"tunes":1578},{"text":1577},"Ground truth is therefore an evaluation construct. A Source of Truth is an application\u002Fdomain authority construct. They can overlap, but they are not interchangeable.",{},{"id":519,"data":1580,"type":42,"tunes":1582},{"text":1581,"level":245},"Source of Truth vs data quality",{},{"id":524,"data":1584,"type":218,"tunes":1586},{"text":1585},"An authoritative source can still contain errors. Authority says which source officially governs the fact; data quality asks whether that source is accurate, complete, timely, consistent and fit for purpose.",{},{"id":529,"data":1588,"type":218,"tunes":1590},{"text":1589},"When an authoritative system is known to be wrong, architecture should record the defect, correction process or exception instead of silently substituting an unofficial source and hiding the discrepancy.",{},{"id":534,"data":1592,"type":42,"tunes":1594},{"text":1593,"level":246},"A practical Source-of-Truth architecture model",{},{"id":539,"data":1596,"type":226,"tunes":1599},{"body":1597,"title":1598,"variant":233},"The model below is a practical synthesis for AI systems. It is not a W3C, NIST or ISO standard. It separates authority, provenance, retrieval and model-facing context because those responsibilities are often incorrectly collapsed.","Proposed architecture model",{},{"id":545,"data":1601,"type":574,"tunes":1628},{"steps":1602,"title":1627,"orientation":573},[1603,1606,1609,1612,1615,1618,1621,1624],{"label":1604,"description":1605},"1. Define the claim type","Identify what the user is actually asking: current state, policy, historical fact, technical specification, user preference, calculation or interpretation.",{"label":1607,"description":1608},"2. Resolve authority","Determine which source or authority class is permitted to define that type of claim for the required scope, version and time.",{"label":1610,"description":1611},"3. Acquire evidence","Read or retrieve the authoritative source and any necessary supporting or conflicting evidence.",{"label":1613,"description":1614},"4. Preserve provenance","Carry source identity, version, timestamp, locator, transformation history and responsibility metadata.",{"label":1616,"description":1617},"5. Build model context","Supply the relevant evidence to the model without discarding authority and applicability metadata.",{"label":1619,"description":1620},"6. Generate or compute","The model may summarize, compare, reason or transform the evidence, but it does not inherit source authority merely by processing it.",{"label":1622,"description":1623},"7. Validate the claim","Check that the answer is supported by the right source and remains inside its scope and validity boundary.",{"label":1625,"description":1626},"8. Preserve contradictions","If authoritative or relevant sources disagree, expose the conflict rather than fabricating false certainty.","Authority-aware AI answer path",{},{"id":577,"data":1630,"type":42,"tunes":1632},{"text":1631,"level":246},"Authority should be explicit, not inferred from similarity",{},{"id":582,"data":1634,"type":218,"tunes":1636},{"text":1635},"One robust implementation pattern is an authority registry or equivalent policy layer that maps claim classes to authoritative source classes. The implementation can be code, metadata, configuration or domain rules; the important property is that authority is deliberate.",{},{"id":587,"data":1638,"type":318,"tunes":1668},{"content":1639,"stretched":43,"withHeadings":14},[1640,1644,1648,1652,1656,1660,1664],[1641,1642,1643],"Claim class","Authority rule","Fallback behavior",[1645,1646,1647],"Current account state","Read live account service \u002F system of record","If unavailable, report that current state cannot be verified.",[1649,1650,1651],"Product documentation","Current approved documentation version","Older version may be shown only with version warning.",[1653,1654,1655],"Implemented software behavior","Relevant deployed release \u002F source artifact","Documentation alone cannot prove deployed behavior.",[1657,1658,1659],"Internal policy","Approved policy repository and active version","Drafts are supporting material, not current authority.",[1661,1662,1663],"External technical standard","Official standards body publication for relevant version","Secondary explanations can clarify but not override the specification.",[1665,1666,1667],"Research claim","Evidence policy appropriate to the domain","Preserve conflicting evidence and confidence rather than force one source.",{},{"id":620,"data":1670,"type":42,"tunes":1672},{"text":1671,"level":246},"Retrieval should use authority as a ranking constraint",{},{"id":625,"data":1674,"type":218,"tunes":1676},{"text":1675},"Semantic relevance answers “Which candidate looks related to this query?” Authority answers “Which candidate is allowed to establish this fact?” A production retrieval system often needs both.",{},{"id":630,"data":1678,"type":218,"tunes":1680},{"text":1679},"A useful sequence is to first constrain the candidate space by identity, tenant, source class, status, version or date, then rank relevant evidence inside the permitted space. If relevance is calculated before critical authorization or authority filters, the pipeline can return a convincing but invalid result.",{},{"id":635,"data":1682,"type":226,"tunes":1685},{"body":1683,"title":1684,"variant":639},"A search engine can rank a stale summary above the current primary source. A language model can prefer the better-written explanation over the official record. Neither behavior changes which source owns the fact.","Highest similarity is not highest authority",{},{"id":642,"data":1687,"type":42,"tunes":1689},{"text":1688,"level":246},"Freshness is part of authority",{},{"id":647,"data":1691,"type":218,"tunes":1693},{"text":1692},"Many Source-of-Truth failures are actually time failures. The correct source was known, but the system used an old snapshot, stale embedding, cached API response or superseded document.",{},{"id":652,"data":1695,"type":218,"tunes":1697},{"text":1696},"An authority rule should therefore include invalidation or refresh semantics where the fact can change. “CRM is authoritative” is incomplete when the application reads a week-old replicated export.",{},{"id":657,"data":1699,"type":42,"tunes":1701},{"text":1700,"level":246},"Derived values need a trace back to authoritative inputs",{},{"id":662,"data":1703,"type":218,"tunes":1705},{"text":1704},"Some important facts are not stored directly. They are computed from authoritative inputs: a risk score, account total, eligibility status or aggregate metric.",{},{"id":667,"data":1707,"type":218,"tunes":1709},{"text":1708},"For derived values, Source-of-Truth architecture should preserve the input authorities, transformation or calculation version and execution time. W3C PROV's distinction between entities, activities and derivations is useful here because it models how one entity was produced from others.",{},{"id":672,"data":1711,"type":226,"tunes":1714},{"body":1712,"title":1713,"variant":274},"A calculated value is only as current and authoritative as its inputs, transformation and validity conditions. Store the lineage needed to recompute or audit it.","Derived truth is conditional",{},{"id":678,"data":1716,"type":42,"tunes":1718},{"text":1717,"level":246},"What happens when authoritative sources disagree?",{},{"id":683,"data":1720,"type":218,"tunes":1722},{"text":1721},"Conflicts are not edge cases in serious knowledge systems. A signed contract may disagree with a CRM field. Production behavior may disagree with documentation. Two primary historical sources may contradict each other. A current policy can conflict with an outdated local copy.",{},{"id":688,"data":1724,"type":218,"tunes":1726},{"text":1725},"The system needs a resolution policy appropriate to the domain. Sometimes one authority clearly outranks the other. Sometimes the newer version supersedes the old one. Sometimes an expert or business owner must adjudicate. And sometimes the correct result is simply: the evidence is unresolved.",{},{"id":693,"data":1728,"type":318,"tunes":1751},{"content":1729,"stretched":43,"withHeadings":14},[1730,1733,1736,1739,1742,1745,1748],[1731,1732],"Conflict type","Typical handling",[1734,1735],"Current vs superseded version","Use current version for present state; retain older version as historical evidence.",[1737,1738],"System of record vs stale replica","Use system of record; flag replication freshness issue.",[1740,1741],"Contract vs CRM transcription","Executed contract governs contractual wording; CRM discrepancy becomes a correction task.",[1743,1744],"Documentation vs deployed behavior","Distinguish intended behavior from observed\u002Fdeployed behavior; do not silently merge them.",[1746,1747],"Two credible primary sources","Preserve both, assess provenance and scope, and represent unresolved disagreement if no governing authority exists.",[1749,1750],"User memory vs current user setting","Use current explicit setting; mark memory as superseded where appropriate.",{},{"id":719,"data":1753,"type":42,"tunes":1755},{"text":1754,"level":246},"Web search is discovery, not automatically evidence",{},{"id":724,"data":1757,"type":218,"tunes":1759},{"text":1758},"Search engines are excellent discovery systems. Search snippets, result ranking and generated summaries are not automatically primary evidence.",{},{"id":729,"data":1761,"type":218,"tunes":1763},{"text":1762},"For claims that require authority, the search result should lead to the original publication, official record, source document, dataset or other appropriate artifact. The result page helps locate the source; it does not inherit the source's authority.",{},{"id":734,"data":1765,"type":42,"tunes":1767},{"text":1766,"level":246},"The language model should not decide authority by itself",{},{"id":739,"data":1769,"type":218,"tunes":1771},{"text":1770},"A model can help classify a question, extract claims or compare evidence, but authority should not depend only on the model's preference. Models optimize generation from context; they do not possess a guaranteed domain-specific registry of which database, document or organization owns each fact.",{},{"id":744,"data":1773,"type":218,"tunes":1775},{"text":1774},"This is why application architecture should encode critical authority rules deterministically where practical. The model may reason within the boundary, but the boundary itself should not be recreated from scratch for every prompt.",{},{"id":749,"data":1777,"type":42,"tunes":1779},{"text":1778,"level":246},"Authority must survive the execution trace",{},{"id":754,"data":1781,"type":218,"tunes":1783},{"text":1782},"If a production answer is important enough to audit, the trace should make it possible to reconstruct which sources were consulted, which version was used, which passage or record supported the claim, which transformations occurred and whether conflicting evidence was available.",{},{"id":759,"data":1785,"type":218,"tunes":1787},{"text":1786},"This aligns with the broader provenance principle in W3C PROV and with NIST AI RMF Playbook guidance to document sources, origins, transformations, dependencies, constraints and metadata.",{},{"id":764,"data":1789,"type":42,"tunes":1791},{"text":1790,"level":246},"Original implementation evidence: Source of Truth Research Engine",{},{"id":769,"data":1793,"type":226,"tunes":1796},{"body":1794,"title":1795,"variant":233},"The Source of Truth Research Engine is my own implementation evidence for an evidence-first architecture. It demonstrates one concrete way to separate discovery, source acquisition, provenance, claims, evidence classes, contradictions and conclusions. It is an implementation pattern, not a universal industry standard.","Original implementation evidence",{},{"id":775,"data":1798,"type":218,"tunes":1800},{"text":1799},"The engine is designed around a traceable pipeline rather than direct AI summarization: research task → search → original source or digital artifact → local snapshot → SHA-256 → source ID → claim → evidence class → relation or contradiction → interpretation → conclusion.",{},{"id":780,"data":1802,"type":218,"tunes":1804},{"text":1803},"Its shared evidence core stores Sources, Artifacts, provenance, Claims, Relations, Contradictions, a Reference Model and an audit trail. Different research modes can share that core while applying different domain methodologies.",{},{"id":785,"data":1806,"type":218,"tunes":1808},{"text":1807},"The architecture deliberately separates discovery from evidence. Search snippets are not treated as evidence, filenames are not treated as content, AI summaries are not treated as primary sources, and semantic similarity is only a discovery signal until a result is traced back to a concrete source and locator.",{},{"id":790,"data":1810,"type":218,"tunes":1812},{"text":1811},"Original files are preserved and local bytes receive SHA-256 identifiers. Contradictions and rejected hypotheses are not silently deleted. New evidence is allowed to change the current reference model while the prior evidence path remains auditable.",{},{"id":795,"data":1814,"type":318,"tunes":1843},{"content":1815,"stretched":43,"withHeadings":14},[1816,1819,1822,1825,1828,1831,1834,1837,1840],[1817,1818],"Implemented rule","Why it matters for Source-of-Truth architecture",[1820,1821],"Search ≠ evidence","Discovery ranking cannot silently become authority.",[1823,1824],"Filename ≠ content","Metadata clues cannot substitute for reading the actual artifact.",[1826,1827],"Local snapshot + SHA-256","Evidence can be tied to exact bytes rather than a mutable remote label.",[1829,1830],"Source ID + exact locator","Claims can be traced back to the concrete evidence location.",[1832,1833],"Claim\u002Fevidence separation","The assertion is not confused with the material supporting it.",[1835,1836],"Contradictions preserved","The system can represent unresolved disagreement rather than overwrite history.",[1838,1839],"Semantic similarity is discovery only","Retrieval relevance is explicitly separated from evidentiary authority.",[1841,1842],"New evidence may update the model","Source-of-Truth state is versioned and revisable rather than treated as immutable dogma.",{},{"id":827,"data":1845,"type":42,"tunes":1847},{"text":1846,"level":246},"Aaasaasa Document & Knowledge Engine: applying the same boundary to enterprise documents",{},{"id":832,"data":1849,"type":218,"tunes":1851},{"text":1850},"The Aaasaasa Document & Knowledge Engine concept extends the same design principle to enterprise documentation: users should be able to search documents, ask source-grounded questions and review collections against explicit criteria while preserving the distinction between what a document states and what the system infers.",{},{"id":837,"data":1853,"type":218,"tunes":1855},{"text":1854},"The important architecture rule is that a universal retrieval core does not make every collection equally authoritative. Contract documents, maintenance records, financial documents and research material need different authority, validation and coverage rules even when they share ingestion and search infrastructure.",{},{"id":842,"data":1857,"type":42,"tunes":1859},{"text":1858,"level":246},"Common Source-of-Truth failure modes",{},{"id":847,"data":1861,"type":318,"tunes":1902},{"content":1862,"stretched":43,"withHeadings":14},[1863,1866,1869,1872,1875,1878,1881,1884,1887,1890,1893,1896,1899],[1864,1865],"Failure mode","What goes wrong",[1867,1868],"The model is treated as the source of truth","Parametric knowledge can be stale, incomplete, unverifiable or outside the application's authoritative scope.",[1870,1871],"Top retrieval result wins automatically","Similarity is mistaken for authority.",[1873,1874],"Vector database becomes authoritative","Derived index records lose the identity and version of the original source.",[1876,1877],"Everything is copied into one knowledge base","Copies obscure ownership, freshness and correction paths.",[1879,1880],"Memory is reused as current state","Old observations silently override the current system of record.",[1882,1883],"No version metadata","The right document is used for the wrong time period.",[1885,1886],"No source locator","A citation exists but the supporting passage or record cannot be verified.",[1888,1889],"Conflicts are overwritten","The system appears consistent by destroying evidence of disagreement.",[1891,1892],"Generated summaries replace originals","A lossy transformation becomes the apparent authority.",[1894,1895],"Authority is global instead of claim-specific","One source is trusted beyond the domain or fact class it actually owns.",[1897,1898],"Web snippet is treated as evidence","Discovery metadata replaces the original publication.",[1900,1901],"Authoritative data is wrong but exceptions are hidden","Operational defects become invisible and cannot be corrected transparently.",{},{"id":891,"data":1904,"type":42,"tunes":1906},{"text":1905,"level":246},"A practical Source-of-Truth decision framework",{},{"id":896,"data":1908,"type":574,"tunes":1938},{"steps":1909,"title":1937,"orientation":573},[1910,1913,1916,1919,1922,1925,1928,1931,1934],{"label":1911,"description":1912},"1. State the claim precisely","Separate current state, historical state, policy, interpretation, prediction and derived calculation.",{"label":1914,"description":1915},"2. Identify the authority owner","Determine the system, document, institution, person or evidence class responsible for that claim type.",{"label":1917,"description":1918},"3. Define scope","Specify tenant, jurisdiction, product, environment, user, document set or other applicability boundary.",{"label":1920,"description":1921},"4. Define time and version","Determine whether the claim requires current state, a historical snapshot or a specific standard\u002Frelease version.",{"label":1923,"description":1924},"5. Preserve provenance","Record source identity, origin, locator, transformations and responsible agents or processes.",{"label":1926,"description":1927},"6. Define retrieval\u002Faccess path","Make sure the application can actually obtain the authoritative information under the correct identity and permissions.",{"label":1929,"description":1930},"7. Define conflict policy","Decide precedence, supersession, adjudication or explicit unresolved-state behavior.",{"label":1932,"description":1933},"8. Define invalidation","Specify when cached, indexed, remembered or derived representations must be refreshed.",{"label":1935,"description":1936},"9. Validate the answer path","Verify that important generated claims can be traced to the intended authority, not merely to a plausible source.","How to decide what should define a claim",{},{"id":929,"data":1940,"type":42,"tunes":1942},{"text":1941,"level":246},"Source-of-Truth architecture checklist",{},{"id":934,"data":1944,"type":318,"tunes":1984},{"content":1945,"stretched":43,"withHeadings":14},[1946,1948,1951,1954,1957,1960,1963,1966,1969,1972,1975,1978,1981],[1443,1947],"Expected answer",[1949,1950],"What exact fact or state is being established?","A claim precise enough to assign authority.",[1952,1953],"Who or what owns that fact?","Named authoritative system, source class or adjudication rule.",[1955,1956],"Is the authority current for this scope?","Tenant, jurisdiction, environment, user or domain boundary is explicit.",[1958,1959],"Is the version\u002Ftime correct?","Current, historical or version-specific applicability is known.",[1961,1962],"Can the source be verified?","Stable identifier, locator or record reference exists.",[1964,1965],"Is provenance preserved?","Origin, transformation and responsibility metadata survive ingestion and retrieval.",[1967,1968],"Can retrieval return non-authoritative material?","If yes, filters or validation distinguish relevance from authority.",[1970,1971],"Can the source change?","Refresh, invalidation or supersession rules exist.",[1973,1974],"Can sources disagree?","Conflict and adjudication behavior is explicit.",[1976,1977],"Can memory become stale?","Volatile state is re-read from the current authority before consequential use.",[1979,1980],"Can a derived answer be reproduced?","Inputs, transformation version and execution conditions are traceable.",[1982,1983],"Can an auditor reconstruct the answer?","Execution evidence preserves the source path for important claims.",{},{"id":977,"data":1986,"type":42,"tunes":1988},{"text":1987,"level":246},"Common misconceptions",{},{"id":982,"data":1990,"type":318,"tunes":2019},{"content":1991,"stretched":43,"withHeadings":14},[1992,1995,1998,2001,2004,2007,2010,2013,2016],[1993,1994],"Misconception","Correction",[1996,1997],"“Source of Truth means one database.”","One database can be authoritative for one domain; complex systems usually have multiple fact-specific authorities.",[1999,2000],"“The newest document is automatically authoritative.”","Recency helps only when the newer artifact is approved and actually supersedes the older one.",[2002,2003],"“RAG solves truth.”","RAG solves retrieval. Authority, provenance, evidence quality and validity remain separate problems.",[2005,2006],"“A citation proves the answer.”","The cited source must actually support the claim, have the right authority and apply to the current scope.",[2008,2009],"“Provenance tells us what is true.”","Provenance tells us origin and derivation; authority and correctness still require domain rules and evaluation.",[2011,2012],"“The system of record is always correct.”","It is authoritative for the operational record, but data-quality defects can still exist and require visible correction.",[2014,2015],"“If several sources agree, the claim is authoritative.”","Agreement increases evidence but does not necessarily establish ownership or applicability.",[2017,2018],"“AI memory can replace repeated reads.”","Only for information whose staleness risk is acceptable; volatile or consequential state should be refreshed from authority.",{},{"id":1014,"data":2021,"type":42,"tunes":2023},{"text":2022,"level":246},"Edge cases",{},{"id":1019,"data":2025,"type":218,"tunes":2027},{"text":2026},"Some questions are interpretive rather than factual. “Which architecture is best?” has no single Source of Truth. The system can retrieve authoritative constraints and evidence, but the final judgment is an inference that should expose assumptions and trade-offs.",{},{"id":1024,"data":2029,"type":218,"tunes":2031},{"text":2030},"Some domains use distributed authority. A scientific conclusion can depend on multiple studies, datasets and replications. A historical conclusion can depend on conflicting primary and secondary evidence. The architecture should represent evidence structure rather than invent a central database that supposedly owns truth.",{},{"id":1029,"data":2033,"type":218,"tunes":2035},{"text":2034},"A user can also be the authority for subjective personal information: preferences, goals, chosen settings or explicit instructions. Even then, newer explicit input can supersede older memory.",{},{"id":1034,"data":2037,"type":218,"tunes":2039},{"text":2038},"External events can invalidate previously authoritative data. A price feed, inventory system or security policy may have been correct when captured but no longer valid. Snapshot provenance preserves what was true then; it does not make the snapshot current forever.",{},{"id":1039,"data":2041,"type":42,"tunes":2043},{"text":2042,"level":246},"Limitations",{},{"id":1044,"data":2045,"type":218,"tunes":2047},{"text":2046},"Source-of-Truth architecture cannot guarantee that an authoritative source is factually correct. It provides accountability, provenance and deterministic ownership boundaries; data-quality and domain-verification processes remain necessary.",{},{"id":1049,"data":2049,"type":218,"tunes":2051},{"text":2050},"Authority can also be contested. Different institutions can legitimately claim authority in different jurisdictions or methodologies. In those situations the system should expose the authority model and disagreement rather than hide it behind a universal “truth score.”",{},{"id":1054,"data":2053,"type":218,"tunes":2055},{"text":2054},"Finally, authority rules require maintenance. Systems, owners, policies, versions and regulations change. A stale authority registry can be as dangerous as no registry at all.",{},{"id":1059,"data":2057,"type":42,"tunes":2059},{"text":2058,"level":246},"What would change this answer?",{},{"id":1064,"data":2061,"type":218,"tunes":2063},{"text":2062},"The specific authority mapping changes with the domain. Banking, healthcare, software delivery, scientific research and historical analysis have different systems of record, evidentiary rules and regulatory obligations.",{},{"id":1069,"data":2065,"type":218,"tunes":2067},{"text":2066},"The implementation also changes with architecture. A small application can encode authority directly in service calls. A larger platform may need registries, source metadata, policy engines, lineage systems or data contracts. The core principle remains the same: do not let retrieval order or model preference silently decide what counts as authoritative.",{},{"id":1074,"data":2069,"type":42,"tunes":2071},{"text":2070,"level":246},"Related canonical knowledge",{},{"id":1079,"data":2073,"type":218,"tunes":2075},{"text":2074},"Source-of-Truth architecture is a prerequisite for later retrieval and governance concepts because retrieval quality alone cannot determine whether evidence is allowed to define the answer.",{},{"id":1084,"data":2077,"type":1090,"tunes":2082},{"url":2078,"title":2079,"excerpt":2080,"ctaLabel":2081},"https:\u002F\u002Fstajic.de\u002Fblog\u002Fwhat-is-rag-the-simplest-explanation-of-how-it-works","What Is RAG? The Simplest Explanation of How It Works","RAG retrieves external knowledge for a model. This foundation explains why retrieval and authoritative truth are separate responsibilities.","Read the RAG foundation",{},{"id":1093,"data":2084,"type":218,"tunes":2086},{"text":2085},"Memory is another adjacent concept. A reliable agent separates remembered information from current authoritative application state.",{},{"id":1098,"data":2088,"type":1090,"tunes":2093},{"url":2089,"title":2090,"excerpt":2091,"ctaLabel":2092},"https:\u002F\u002Fstajic.de\u002Fblog\u002Fai-agent-memory-is-not-rag-how-to-separate-memory-retrieval-state-and-context","AI Agent Memory Is Not RAG: How to Separate Memory, Retrieval, State and Context","A four-layer architecture separating persistent memory, authoritative state, retrieval and the context actually supplied to the model.","Read the memory architecture article",{},{"id":1106,"data":2095,"type":218,"tunes":2097},{"text":2096},"Authority also connects directly to answer validity. Even an authoritative source supports only claims inside its version, date, scope and evidence boundary.",{},{"id":1111,"data":2099,"type":1090,"tunes":2104},{"url":2100,"title":2101,"excerpt":2102,"ctaLabel":2103},"https:\u002F\u002Fstajic.de\u002Fblog\u002Fthe-answer-validity-boundary-the-missing-layer-between-relevance-and-reliable-ai-answers","The Answer Validity Boundary: The Missing Layer Between Relevance and Reliable AI Answers","A framework for making explicit where an AI answer remains supported by evidence and which changed conditions invalidate that support.","Read the Answer Validity Boundary",{},{"id":1119,"data":2106,"type":42,"tunes":2108},{"text":2107,"level":246},"Frequently asked questions",{},{"id":1124,"data":2110,"type":1124,"tunes":2137},{"items":2111,"title":2136},[2112,2115,2118,2121,2124,2127,2130,2133],{"id":1128,"answer":2113,"question":2114},"It is the authoritative source or authority rule that determines which source is allowed to establish a specific fact, state or rule for a defined scope, version and time.","What is a Source of Truth in an AI system?",{"id":1132,"answer":2116,"question":2117},"Normally no. A language model can generate, summarize and reason, but its parametric knowledge is not automatically authoritative for current application state, company policy, a specific document version or a regulated domain fact.","Is the language model a Source of Truth?",{"id":1136,"answer":2119,"question":2120},"Not automatically. A vector database is usually an index or retrieval store. It should preserve references to the authoritative original source and version rather than silently replacing them.","Is a vector database the Source of Truth for RAG?",{"id":1140,"answer":2122,"question":2123},"Provenance describes where data came from, how it was produced or transformed and who or what was involved. Source-of-Truth rules determine whether that source has authority for the specific claim.","What is the difference between provenance and Source of Truth?",{"id":1144,"answer":2125,"question":2126},"Yes. In complex systems this is normal because different facts belong to different authoritative systems or source classes.","Can an AI system have multiple Sources of Truth?",{"id":1148,"answer":2128,"question":2129},"The system needs a domain-specific conflict rule: precedence, version supersession, expert adjudication or an explicit unresolved state. It should not silently choose the source the model prefers.","What happens when two authoritative sources disagree?",{"id":1152,"answer":2131,"question":2132},"No. RAG retrieves candidates. Authority-aware metadata, filters, source policy and validation are needed to ensure that important claims use the correct source.","Does RAG guarantee that an AI answer uses the Source of Truth?",{"id":1156,"answer":2134,"question":2135},"Yes. Authority and correctness are different properties. An authoritative system can contain a data-quality defect, which should be corrected transparently rather than hidden by substituting an unofficial source.","Can the Source of Truth be wrong?","Source of Truth in AI systems",{},{"id":1162,"data":2139,"type":42,"tunes":2141},{"text":2140,"level":246},"Glossary",{},{"id":1167,"data":2143,"type":1167,"tunes":2176},{"title":2144,"entries":2145},"Key Source-of-Truth terms",[2146,2149,2152,2155,2158,2161,2164,2167,2170,2173],{"term":2147,"anchor":1173,"definition":2148},"Source of Truth","The authoritative source or rule permitted to establish a particular fact, state or rule for a defined scope, version and time.",{"term":2150,"anchor":1177,"definition":2151},"System of record","The authoritative operational system responsible for a defined class of records or current business state.",{"term":2153,"anchor":1181,"definition":2154},"Provenance","Information describing the origin, derivation, transformations, responsible agents and history of data or another entity.",{"term":2156,"anchor":1185,"definition":2157},"Evidence","Information or an artifact that supports, contradicts or constrains a claim.",{"term":2159,"anchor":327,"definition":2160},"Authority","The application or domain rule determining which source is entitled to define a specific claim.",{"term":2162,"anchor":1192,"definition":2163},"Freshness","Whether a representation remains current enough for the claim or operation in which it is used.",{"term":2165,"anchor":1196,"definition":2166},"Supersession","The explicit replacement of an older authoritative version by a newer one while preserving historical traceability.",{"term":2168,"anchor":1200,"definition":2169},"Ground truth","A reference answer, label or outcome used to evaluate a system; it is an evaluation construct rather than automatically the application's Source of Truth.",{"term":2171,"anchor":1204,"definition":2172},"Lineage","The trace of how data or derived values flow and transform across sources and processing steps.",{"term":2174,"anchor":1208,"definition":2175},"Validity boundary","The conditions of scope, time, version, evidence and assumptions inside which a claim remains supported.",{},{"id":1212,"data":2178,"type":42,"tunes":2180},{"text":2179,"level":246},"Conclusion",{},{"id":1217,"data":2182,"type":218,"tunes":2184},{"text":2183},"Reliable AI does not come from giving the model more information. It comes from knowing which information is allowed to define the claim, preserving where that information came from, retrieving the correct version and keeping the final answer inside the source's scope.",{},{"id":1222,"data":2186,"type":218,"tunes":2188},{"text":2187},"That is why Source of Truth, provenance, retrieval, memory and context must remain separate concepts. The Source of Truth defines authority. Provenance explains origin. Retrieval finds candidates. Memory preserves selected past information. Context is what the model sees. Generation turns those inputs into an output.",{},{"id":1227,"data":2190,"type":218,"tunes":2192},{"text":2191},"When those layers remain explicit, an AI system can do more than sound plausible: important claims can be traced back to the source that actually had the right to establish them.",{},{"id":1232,"data":2194,"type":42,"tunes":2196},{"text":2195,"level":246},"Primary sources and implementation evidence",{},{"id":1237,"data":2198,"type":218,"tunes":2200},{"text":2199},"The external sources below support provenance and AI risk-management claims. The Source of Truth Research Engine section is original implementation evidence and is explicitly presented as one implementation pattern rather than a universal standard.",{},{"id":1242,"data":2202,"type":1249,"tunes":2207},{"link":1244,"meta":2203},{"image":2204,"title":2205,"description":2206},{"url":304},"W3C PROV-DM — The PROV Data Model","W3C Recommendation defining a domain-agnostic provenance model around entities, activities, agents, derivations and responsibility.",{},{"id":1252,"data":2209,"type":1249,"tunes":2214},{"link":1254,"meta":2210},{"image":2211,"title":2212,"description":2213},{"url":304},"W3C Provenance Working Group — Publications","Official index of W3C PROV Recommendations and related specifications for provenance interchange and constraints.",{},{"id":1261,"data":2216,"type":1249,"tunes":2221},{"link":1263,"meta":2217},{"image":2218,"title":2219,"description":2220},{"url":304},"NIST AI Risk Management Framework","NIST's voluntary framework for incorporating trustworthiness and risk-management considerations across the AI lifecycle; AI RMF 1.0 is currently being revised.",{},{"id":1270,"data":2223,"type":1249,"tunes":2228},{"link":1272,"meta":2224},{"image":2225,"title":2226,"description":2227},{"url":304},"NIST AI RMF Playbook","Operational guidance aligned to the AI RMF, including documentation practices for data provenance, sources, origins, transformations, dependencies, constraints and metadata.",{},{"id":1279,"data":2230,"type":1249,"tunes":2235},{"link":1281,"meta":2231},{"image":2232,"title":2233,"description":2234},{"url":304},"NIST AI RMF Playbook — Measure","Guidance on documenting measurement, data provenance and contextual interpretation of AI system outputs.",{},{"id":1288,"data":2237,"type":1249,"tunes":2242},{"link":1290,"meta":2238},{"image":2239,"title":2240,"description":2241},{"url":304},"NIST AI 600-1 — Generative AI Profile","NIST generative-AI profile, including provenance and information-integrity considerations for generative AI systems.",{},"2.31.6","A Source of Truth defines which source is authoritative for a specific fact or state. Learn how it differs from RAG, provenance, memory, context, vector databases and systems of record.",{"lang":7,"title":208,"content":210,"contentJson":2246,"excerpt":1297},{"time":212,"blocks":2247,"version":1296},[2248,2251,2254,2257,2260,2263,2266,2269,2272,2275,2278,2281,2284,2287,2290,2306,2309,2312,2315,2318,2330,2333,2336,2339,2342,2345,2348,2351,2354,2357,2360,2363,2366,2369,2372,2375,2378,2381,2384,2387,2390,2393,2396,2399,2402,2405,2408,2411,2414,2417,2420,2423,2435,2438,2441,2452,2455,2458,2461,2464,2467,2470,2473,2476,2479,2482,2485,2488,2491,2494,2505,2508,2511,2514,2517,2520,2523,2526,2529,2532,2535,2538,2541,2544,2547,2550,2563,2566,2569,2572,2575,2592,2595,2608,2611,2628,2631,2644,2647,2650,2653,2656,2659,2662,2665,2668,2671,2674,2677,2680,2683,2686,2689,2692,2695,2698,2701,2704,2716,2719,2733,2736,2739,2742,2745,2748,2751,2756,2761,2766,2771,2776],{"id":215,"data":2249,"type":218,"tunes":2250},{"text":217},{},{"id":221,"data":2252,"type":226,"tunes":2253},{"body":223,"title":224,"variant":225},{},{"id":229,"data":2255,"type":226,"tunes":2256},{"body":231,"title":232,"variant":233},{},{"id":236,"data":2258,"type":226,"tunes":2259},{"body":238,"title":239,"variant":233},{},{"id":242,"data":2261,"type":247,"tunes":2262},{"title":244,"maxLevel":245,"minLevel":246},{},{"id":250,"data":2264,"type":42,"tunes":2265},{"text":252,"level":246},{},{"id":255,"data":2267,"type":218,"tunes":2268},{"text":257},{},{"id":260,"data":2270,"type":218,"tunes":2271},{"text":262},{},{"id":265,"data":2273,"type":218,"tunes":2274},{"text":267},{},{"id":270,"data":2276,"type":226,"tunes":2277},{"body":272,"title":273,"variant":274},{},{"id":277,"data":2279,"type":42,"tunes":2280},{"text":279,"level":246},{},{"id":282,"data":2282,"type":218,"tunes":2283},{"text":284},{},{"id":287,"data":2285,"type":218,"tunes":2286},{"text":289},{},{"id":292,"data":2288,"type":218,"tunes":2289},{"text":294},{},{"id":297,"data":2291,"type":329,"tunes":2305},{"rows":2292,"title":317,"layout":318,"columns":2301},[2293,2295,2297,2299],{"id":301,"label":302,"values":2294},[304,304,304],{"id":306,"label":307,"values":2296},[304,304,304],{"id":310,"label":311,"values":2298},[304,304,304],{"id":314,"label":315,"values":2300},[304,304,304],[2302,2303,2304],{"id":321,"label":322},{"id":324,"label":325},{"id":327,"label":328},{},{"id":332,"data":2307,"type":42,"tunes":2308},{"text":334,"level":246},{},{"id":337,"data":2310,"type":218,"tunes":2311},{"text":339},{},{"id":342,"data":2313,"type":218,"tunes":2314},{"text":344},{},{"id":347,"data":2316,"type":42,"tunes":2317},{"text":349,"level":246},{},{"id":352,"data":2319,"type":318,"tunes":2329},{"content":2320,"stretched":43,"withHeadings":14},[2321,2322,2323,2324,2325,2326,2327,2328],[356,357,358],[360,361,362],[364,365,366],[368,369,370],[372,373,374],[376,377,378],[380,381,382],[384,385,386],{},{"id":389,"data":2331,"type":218,"tunes":2332},{"text":391},{},{"id":394,"data":2334,"type":42,"tunes":2335},{"text":396,"level":246},{},{"id":399,"data":2337,"type":42,"tunes":2338},{"text":401,"level":245},{},{"id":404,"data":2340,"type":218,"tunes":2341},{"text":406},{},{"id":409,"data":2343,"type":218,"tunes":2344},{"text":411},{},{"id":414,"data":2346,"type":42,"tunes":2347},{"text":416,"level":245},{},{"id":419,"data":2349,"type":218,"tunes":2350},{"text":421},{},{"id":424,"data":2352,"type":218,"tunes":2353},{"text":426},{},{"id":429,"data":2355,"type":42,"tunes":2356},{"text":431,"level":245},{},{"id":434,"data":2358,"type":218,"tunes":2359},{"text":436},{},{"id":439,"data":2361,"type":218,"tunes":2362},{"text":441},{},{"id":444,"data":2364,"type":42,"tunes":2365},{"text":446,"level":245},{},{"id":449,"data":2367,"type":218,"tunes":2368},{"text":451},{},{"id":454,"data":2370,"type":218,"tunes":2371},{"text":456},{},{"id":459,"data":2373,"type":42,"tunes":2374},{"text":461,"level":245},{},{"id":464,"data":2376,"type":218,"tunes":2377},{"text":466},{},{"id":469,"data":2379,"type":218,"tunes":2380},{"text":471},{},{"id":474,"data":2382,"type":42,"tunes":2383},{"text":476,"level":245},{},{"id":479,"data":2385,"type":218,"tunes":2386},{"text":481},{},{"id":484,"data":2388,"type":218,"tunes":2389},{"text":486},{},{"id":489,"data":2391,"type":42,"tunes":2392},{"text":491,"level":245},{},{"id":494,"data":2394,"type":218,"tunes":2395},{"text":496},{},{"id":499,"data":2397,"type":218,"tunes":2398},{"text":501},{},{"id":504,"data":2400,"type":42,"tunes":2401},{"text":506,"level":245},{},{"id":509,"data":2403,"type":218,"tunes":2404},{"text":511},{},{"id":514,"data":2406,"type":218,"tunes":2407},{"text":516},{},{"id":519,"data":2409,"type":42,"tunes":2410},{"text":521,"level":245},{},{"id":524,"data":2412,"type":218,"tunes":2413},{"text":526},{},{"id":529,"data":2415,"type":218,"tunes":2416},{"text":531},{},{"id":534,"data":2418,"type":42,"tunes":2419},{"text":536,"level":246},{},{"id":539,"data":2421,"type":226,"tunes":2422},{"body":541,"title":542,"variant":233},{},{"id":545,"data":2424,"type":574,"tunes":2434},{"steps":2425,"title":572,"orientation":573},[2426,2427,2428,2429,2430,2431,2432,2433],{"label":549,"description":550},{"label":552,"description":553},{"label":555,"description":556},{"label":558,"description":559},{"label":561,"description":562},{"label":564,"description":565},{"label":567,"description":568},{"label":570,"description":571},{},{"id":577,"data":2436,"type":42,"tunes":2437},{"text":579,"level":246},{},{"id":582,"data":2439,"type":218,"tunes":2440},{"text":584},{},{"id":587,"data":2442,"type":318,"tunes":2451},{"content":2443,"stretched":43,"withHeadings":14},[2444,2445,2446,2447,2448,2449,2450],[591,592,593],[595,596,597],[599,600,601],[603,604,605],[607,608,609],[611,612,613],[615,616,617],{},{"id":620,"data":2453,"type":42,"tunes":2454},{"text":622,"level":246},{},{"id":625,"data":2456,"type":218,"tunes":2457},{"text":627},{},{"id":630,"data":2459,"type":218,"tunes":2460},{"text":632},{},{"id":635,"data":2462,"type":226,"tunes":2463},{"body":637,"title":638,"variant":639},{},{"id":642,"data":2465,"type":42,"tunes":2466},{"text":644,"level":246},{},{"id":647,"data":2468,"type":218,"tunes":2469},{"text":649},{},{"id":652,"data":2471,"type":218,"tunes":2472},{"text":654},{},{"id":657,"data":2474,"type":42,"tunes":2475},{"text":659,"level":246},{},{"id":662,"data":2477,"type":218,"tunes":2478},{"text":664},{},{"id":667,"data":2480,"type":218,"tunes":2481},{"text":669},{},{"id":672,"data":2483,"type":226,"tunes":2484},{"body":674,"title":675,"variant":274},{},{"id":678,"data":2486,"type":42,"tunes":2487},{"text":680,"level":246},{},{"id":683,"data":2489,"type":218,"tunes":2490},{"text":685},{},{"id":688,"data":2492,"type":218,"tunes":2493},{"text":690},{},{"id":693,"data":2495,"type":318,"tunes":2504},{"content":2496,"stretched":43,"withHeadings":14},[2497,2498,2499,2500,2501,2502,2503],[697,698],[700,701],[703,704],[706,707],[709,710],[712,713],[715,716],{},{"id":719,"data":2506,"type":42,"tunes":2507},{"text":721,"level":246},{},{"id":724,"data":2509,"type":218,"tunes":2510},{"text":726},{},{"id":729,"data":2512,"type":218,"tunes":2513},{"text":731},{},{"id":734,"data":2515,"type":42,"tunes":2516},{"text":736,"level":246},{},{"id":739,"data":2518,"type":218,"tunes":2519},{"text":741},{},{"id":744,"data":2521,"type":218,"tunes":2522},{"text":746},{},{"id":749,"data":2524,"type":42,"tunes":2525},{"text":751,"level":246},{},{"id":754,"data":2527,"type":218,"tunes":2528},{"text":756},{},{"id":759,"data":2530,"type":218,"tunes":2531},{"text":761},{},{"id":764,"data":2533,"type":42,"tunes":2534},{"text":766,"level":246},{},{"id":769,"data":2536,"type":226,"tunes":2537},{"body":771,"title":772,"variant":233},{},{"id":775,"data":2539,"type":218,"tunes":2540},{"text":777},{},{"id":780,"data":2542,"type":218,"tunes":2543},{"text":782},{},{"id":785,"data":2545,"type":218,"tunes":2546},{"text":787},{},{"id":790,"data":2548,"type":218,"tunes":2549},{"text":792},{},{"id":795,"data":2551,"type":318,"tunes":2562},{"content":2552,"stretched":43,"withHeadings":14},[2553,2554,2555,2556,2557,2558,2559,2560,2561],[799,800],[802,803],[805,806],[808,809],[811,812],[814,815],[817,818],[820,821],[823,824],{},{"id":827,"data":2564,"type":42,"tunes":2565},{"text":829,"level":246},{},{"id":832,"data":2567,"type":218,"tunes":2568},{"text":834},{},{"id":837,"data":2570,"type":218,"tunes":2571},{"text":839},{},{"id":842,"data":2573,"type":42,"tunes":2574},{"text":844,"level":246},{},{"id":847,"data":2576,"type":318,"tunes":2591},{"content":2577,"stretched":43,"withHeadings":14},[2578,2579,2580,2581,2582,2583,2584,2585,2586,2587,2588,2589,2590],[851,852],[854,855],[857,858],[860,861],[863,864],[866,867],[869,870],[872,873],[875,876],[878,879],[881,882],[884,885],[887,888],{},{"id":891,"data":2593,"type":42,"tunes":2594},{"text":893,"level":246},{},{"id":896,"data":2596,"type":574,"tunes":2607},{"steps":2597,"title":926,"orientation":573},[2598,2599,2600,2601,2602,2603,2604,2605,2606],{"label":900,"description":901},{"label":903,"description":904},{"label":906,"description":907},{"label":909,"description":910},{"label":912,"description":913},{"label":915,"description":916},{"label":918,"description":919},{"label":921,"description":922},{"label":924,"description":925},{},{"id":929,"data":2609,"type":42,"tunes":2610},{"text":931,"level":246},{},{"id":934,"data":2612,"type":318,"tunes":2627},{"content":2613,"stretched":43,"withHeadings":14},[2614,2615,2616,2617,2618,2619,2620,2621,2622,2623,2624,2625,2626],[356,938],[940,941],[943,944],[946,947],[949,950],[952,953],[955,956],[958,959],[961,962],[964,965],[967,968],[970,971],[973,974],{},{"id":977,"data":2629,"type":42,"tunes":2630},{"text":979,"level":246},{},{"id":982,"data":2632,"type":318,"tunes":2643},{"content":2633,"stretched":43,"withHeadings":14},[2634,2635,2636,2637,2638,2639,2640,2641,2642],[986,987],[989,990],[992,993],[995,996],[998,999],[1001,1002],[1004,1005],[1007,1008],[1010,1011],{},{"id":1014,"data":2645,"type":42,"tunes":2646},{"text":1016,"level":246},{},{"id":1019,"data":2648,"type":218,"tunes":2649},{"text":1021},{},{"id":1024,"data":2651,"type":218,"tunes":2652},{"text":1026},{},{"id":1029,"data":2654,"type":218,"tunes":2655},{"text":1031},{},{"id":1034,"data":2657,"type":218,"tunes":2658},{"text":1036},{},{"id":1039,"data":2660,"type":42,"tunes":2661},{"text":1041,"level":246},{},{"id":1044,"data":2663,"type":218,"tunes":2664},{"text":1046},{},{"id":1049,"data":2666,"type":218,"tunes":2667},{"text":1051},{},{"id":1054,"data":2669,"type":218,"tunes":2670},{"text":1056},{},{"id":1059,"data":2672,"type":42,"tunes":2673},{"text":1061,"level":246},{},{"id":1064,"data":2675,"type":218,"tunes":2676},{"text":1066},{},{"id":1069,"data":2678,"type":218,"tunes":2679},{"text":1071},{},{"id":1074,"data":2681,"type":42,"tunes":2682},{"text":1076,"level":246},{},{"id":1079,"data":2684,"type":218,"tunes":2685},{"text":1081},{},{"id":1084,"data":2687,"type":1090,"tunes":2688},{"url":1086,"title":1087,"excerpt":1088,"ctaLabel":1089},{},{"id":1093,"data":2690,"type":218,"tunes":2691},{"text":1095},{},{"id":1098,"data":2693,"type":1090,"tunes":2694},{"url":1100,"title":1101,"excerpt":1102,"ctaLabel":1103},{},{"id":1106,"data":2696,"type":218,"tunes":2697},{"text":1108},{},{"id":1111,"data":2699,"type":1090,"tunes":2700},{"url":1113,"title":1114,"excerpt":1115,"ctaLabel":1116},{},{"id":1119,"data":2702,"type":42,"tunes":2703},{"text":1121,"level":246},{},{"id":1124,"data":2705,"type":1124,"tunes":2715},{"items":2706,"title":1159},[2707,2708,2709,2710,2711,2712,2713,2714],{"id":1128,"answer":1129,"question":1130},{"id":1132,"answer":1133,"question":1134},{"id":1136,"answer":1137,"question":1138},{"id":1140,"answer":1141,"question":1142},{"id":1144,"answer":1145,"question":1146},{"id":1148,"answer":1149,"question":1150},{"id":1152,"answer":1153,"question":1154},{"id":1156,"answer":1157,"question":1158},{},{"id":1162,"data":2717,"type":42,"tunes":2718},{"text":1164,"level":246},{},{"id":1167,"data":2720,"type":1167,"tunes":2732},{"title":1169,"entries":2721},[2722,2723,2724,2725,2726,2727,2728,2729,2730,2731],{"term":1172,"anchor":1173,"definition":1174},{"term":1176,"anchor":1177,"definition":1178},{"term":1180,"anchor":1181,"definition":1182},{"term":1184,"anchor":1185,"definition":1186},{"term":1188,"anchor":327,"definition":1189},{"term":1191,"anchor":1192,"definition":1193},{"term":1195,"anchor":1196,"definition":1197},{"term":1199,"anchor":1200,"definition":1201},{"term":1203,"anchor":1204,"definition":1205},{"term":1207,"anchor":1208,"definition":1209},{},{"id":1212,"data":2734,"type":42,"tunes":2735},{"text":1214,"level":246},{},{"id":1217,"data":2737,"type":218,"tunes":2738},{"text":1219},{},{"id":1222,"data":2740,"type":218,"tunes":2741},{"text":1224},{},{"id":1227,"data":2743,"type":218,"tunes":2744},{"text":1229},{},{"id":1232,"data":2746,"type":42,"tunes":2747},{"text":1234,"level":246},{},{"id":1237,"data":2749,"type":218,"tunes":2750},{"text":1239},{},{"id":1242,"data":2752,"type":1249,"tunes":2755},{"link":1244,"meta":2753},{"image":2754,"title":1247,"description":1248},{"url":304},{},{"id":1252,"data":2757,"type":1249,"tunes":2760},{"link":1254,"meta":2758},{"image":2759,"title":1257,"description":1258},{"url":304},{},{"id":1261,"data":2762,"type":1249,"tunes":2765},{"link":1263,"meta":2763},{"image":2764,"title":1266,"description":1267},{"url":304},{},{"id":1270,"data":2767,"type":1249,"tunes":2770},{"link":1272,"meta":2768},{"image":2769,"title":1275,"description":1276},{"url":304},{},{"id":1279,"data":2772,"type":1249,"tunes":2775},{"link":1281,"meta":2773},{"image":2774,"title":1284,"description":1285},{"url":304},{},{"id":1288,"data":2777,"type":1249,"tunes":2780},{"link":1290,"meta":2778},{"image":2779,"title":1293,"description":1294},{"url":304},{},"Post erfolgreich abgerufen",{"items":2783,"source":2868,"manualIds":2869,"manualMatchedIds":2870},[2784,2791,2798,2805,2812,2819,2826,2833,2840,2847,2854,2861],{"id":2785,"slug":2786,"title":2787,"excerpt":2788,"featuredImage":2789,"publishedAt":2790},"494","air-gapped-ai-how-ai-systems-work-without-internet-or-cloud-access","Air-Gapped AI: Kako AI sistemi funkcionišu bez interneta ili pristupa oblaku","Air-gapped AI pokreće modele, RAG i AI aplikacije unutar izolovanog bezbednosnog domena bez internet ili cloud zavisnosti. Saznajte kako modeli, podaci, ažuriranja i alati funkcionišu offline.","\u002Fuploads\u002F2026\u002F10\u002Fair-gapped-ai-how-ai-systems-work-without-internet-or-cloud-access-1791487983978-e6xqf0.webp","2026-10-08T11:32:00.000Z",{"id":2792,"slug":2793,"title":2794,"excerpt":2795,"featuredImage":2796,"publishedAt":2797},"490","rbac-vs-tenant-isolation-two-different-security-boundaries","RBAC naspram izolacije zakupaca: dve različite bezbednosne granice","RBAC kontroliše šta korisnik sme da radi; izolacija zakupaca kontroliše kojim resursima tog zakupca ta radnja može da pristupi. Saznajte zašto bezbednost višekorisničkog SaaS-a zahteva obe granice.","\u002Fuploads\u002F2026\u002F10\u002Frbac-vs-tenant-isolation-two-different-security-boundaries-1791485111528-qqtzby.webp","2026-10-08T14:43:00.000Z",{"id":2799,"slug":2800,"title":2801,"excerpt":2802,"featuredImage":2803,"publishedAt":2804},"468","ai-agent-memory-is-not-rag-how-to-separate-memory-retrieval-state-and-context","Memorija AI agenta nije RAG: Kako razdvojiti memoriju, pronalaženje, stanje i kontekst","Memorija agenta, RAG, stanje i kontekst često se koriste kao da su međusobno zamenjivi. Oni to nisu. Ovaj praktični arhitektonski model razdvaja ova četiri sloja, pokazuje gde svaki pripada i objašnjava šta se kvari kada ih sistemi stope u jedno.","\u002Fuploads\u002F2026\u002F09\u002Fai-agent-memory-is-not-rag-how-to-separate-memory-retrieval-state-and-context-1790350560308-np0xy6.webp","2026-09-25T11:34:00.000Z",{"id":2806,"slug":2807,"title":2808,"excerpt":2809,"featuredImage":2810,"publishedAt":2811},"484","what-is-an-ai-platform-architect-models-data-runtime-security-and-operations","Šta je arhitekta AI platforme? Modeli, podaci, izvršno okruženje, bezbednost i operacije","Arhitekta AI platforme projektuje višekratno upotrebljive AI temelje kroz modele, provajdere, pretragu, agente, identitet, bezbednost, evaluaciju, opservabilnost i operacije.","\u002Fuploads\u002F2026\u002F10\u002Fwhat-is-an-ai-platform-architect-models-data-runtime-security-and-operations-1791477229171-ou3zcc.webp","2026-10-08T12:32:00.000Z",{"id":2813,"slug":2814,"title":2815,"excerpt":2816,"featuredImage":2817,"publishedAt":2818},"466","the-gpu-is-not-the-product-future-proof-private-ai-architecture","GPU nije proizvod: Privatna AI arhitektura spremna za budućnost","Privatna AI infrastruktura ne bi trebalo da bude projektovana oko jednog GPU-a ili jednog modela. Otporniji pristup kombinuje brze GPU-ove za inferenciju, memorijski bogate AI sisteme, čvorove za fizički AI i opcione vodeće modele u oblaku iza sloja za rutiranje koji prepoznaje mogućnosti.","\u002Fuploads\u002F2026\u002F09\u002Fthe-gpu-is-not-the-product-future-proof-private-ai-architecture-1790140878812-8hsl39.webp","2026-09-23T01:19:00.000Z",{"id":2820,"slug":2821,"title":2822,"excerpt":2823,"featuredImage":2824,"publishedAt":2825},"489","agentic-ai-explained-when-an-ai-system-can-plan-use-tools-and-act","Agentna AI objašnjena: Kada AI sistem može da planira, koristi alate i deluje","Agentna AI koristi modele unutar višekoračnih izvršnih petlji gde mogu da biraju alate, posmatraju rezultate, ažuriraju stanje i prilagode svoju sledeću akciju unutar eksplicitnih granica izvršavanja i dozvola.","\u002Fuploads\u002F2026\u002F10\u002Fagentic-ai-explained-when-an-ai-system-can-plan-use-tools-and-act-1791481499084-wnji2a.webp","2026-10-08T11:43:00.000Z",{"id":2827,"slug":2828,"title":2829,"excerpt":2830,"featuredImage":2831,"publishedAt":2832},"479","where-does-an-llm-get-its-data-rag-data-sources-in-python","Odakle LLM dobija svoje podatke? RAG izvori podataka u Python-u","LLM ne zna magično vaše fajlove, baze podataka ili API-je. Ovaj praktični nastavak RAG serije pokazuje, uz jednostavan Python, kako eksterni podaci postaju dokazi koji se mogu pronaći: od tekstualnih fajlova i SQL-a do pretrage punog teksta, embeddinga, sastavljanja konteksta i konačnog LLM poziva.","\u002Fuploads\u002F2026\u002F09\u002Fwhere-does-an-llm-get-its-data-rag-data-sources-in-python-1790517200521-nfsi5i.webp","2026-09-27T05:51:00.000Z",{"id":2834,"slug":2835,"title":2836,"excerpt":2837,"featuredImage":2838,"publishedAt":2839},"485","enterprise-ai-architecture-what-changes-when-ai-enters-a-company","Enterprise AI arhitektura: Šta se menja kada AI uđe u kompaniju","Enterprise AI arhitektura objašnjava kako AI menja korporativne sisteme kroz autoritet podataka, identitet, dozvole, provajdere, rizik, upravljanje, evaluaciju, usklađenost i operacije.","\u002Fuploads\u002F2026\u002F10\u002Fenterprise-ai-architecture-what-changes-when-ai-enters-a-company-1791478161363-czrwaq.webp","2026-10-08T10:48:00.000Z",{"id":2841,"slug":2842,"title":2843,"excerpt":2844,"featuredImage":2845,"publishedAt":2846},"483","what-is-an-ai-solution-architect-system-boundaries-responsibilities-and-trade-offs","Šta je AI rešenje arhitekta? Granice sistema, odgovornosti i kompromisi","AI Solution Architect pretvara poslovne zahteve u AI sistem spreman za produkciju, obuhvatajući podatke, modele, alate, bezbednost, izvršno okruženje, evaluaciju i operacije.","\u002Fuploads\u002F2026\u002F10\u002Fwhat-is-an-ai-solution-architect-system-boundaries-responsibilities-and-trade-offs-1791476643267-1st5xz.webp","2026-10-08T12:23:00.000Z",{"id":2848,"slug":2849,"title":2850,"excerpt":2851,"featuredImage":2852,"publishedAt":2853},"471","how-to-know-whether-an-ai-agent-actually-used-the-right-evidence","Kako znati da li je AI agent zaista koristio prave dokaze","AI agent može citirati izvore i ipak koristiti pogrešne dokaze. Ovaj članak predstavlja praktičnu metodu za proveru potkrepljenosti tvrdnji, autoriteta izvora, primenjivosti, porekla i toga da li su dokazi zaista uticali na odgovor.","\u002Fuploads\u002F2026\u002F09\u002Fhow-to-know-whether-an-ai-agent-actually-used-the-right-evidence-1790351317188-o5z9ve.webp","2026-09-25T11:47:00.000Z",{"id":2855,"slug":2856,"title":2857,"excerpt":2858,"featuredImage":2859,"publishedAt":2860},"481","generative-ai-explained-models-retrieval-tools-and-applications-are-not-the-same-thing","Generativna veštačka inteligencija objašnjena: modeli, pretraga, alati i aplikacije nisu ista stvar","Generativna AI je više od modela. Saznajte kako se modeli, pretraga, alati, kontekst, okruženja i aplikacije uklapaju u produkcione AI sisteme.","\u002Fuploads\u002F2026\u002F10\u002Fgenerative-ai-explained-models-retrieval-tools-and-applications-are-not-the-same-thing-1791475411822-pp0dvz.webp","2026-10-08T12:00:00.000Z",{"id":2862,"slug":2863,"title":2864,"excerpt":2865,"featuredImage":2866,"publishedAt":2867},"476","mcp-vs-a2a-vs-ucp-vs-ap2-vs-a2ui-the-agent-protocol-stack-explained","MCP vs A2A vs UCP vs AP2 vs A2UI: Objašnjen stek agentskih protokola","MCP, A2A, UCP, AP2 i A2UI se često predstavljaju kao konkurentski standardi za agente. Oni uglavnom rešavaju različite probleme interoperabilnosti. Ovaj vodič mapira svaki protokol na granicu koju zapravo standardizuje—i pokazuje kako oni mogu da rade zajedno u jednom produkcionom sistemu.","\u002Fuploads\u002F2026\u002F09\u002Fmcp-vs-a2a-vs-ucp-vs-ap2-vs-a2ui-the-agent-protocol-stack-explained-1790352625869-2ezle0.webp","2026-09-25T12:09:00.000Z","fallback",[],[]]