{"id":12519,"date":"2024-05-22T18:16:07","date_gmt":"2024-05-22T18:16:07","guid":{"rendered":"https:\/\/dailyai.com\/?p=12519"},"modified":"2024-05-24T10:16:44","modified_gmt":"2024-05-24T10:16:44","slug":"decoding-the-ai-mind-anthropic-researchers-peer-inside-the-black-box","status":"publish","type":"post","link":"https:\/\/dailyai.com\/nb\/2024\/05\/decoding-the-ai-mind-anthropic-researchers-peer-inside-the-black-box\/","title":{"rendered":"Avkoding av AI-sinnet: Antropiske forskere kikker inn i den \"svarte boksen\""},"content":{"rendered":"<p><b><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\">Anthropic<\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/span> forskere har lykkes med \u00e5 identifisere millioner av konsepter innenfor <span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\">Claude<\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/span> Sonnet, en av deres avanserte LLM-er.<\/b><\/p>\n<p><span style=\"font-weight: 400;\">AI-modeller anses ofte som svarte bokser, noe som betyr at man ikke kan \"se\" inn i dem for \u00e5 forst\u00e5 n\u00f8yaktig hvordan de fungerer.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">N\u00e5r du gir en LLM en input, genererer den et svar, men det er uklart hva som ligger bak valgene. <\/span><\/p>\n<p><span style=\"font-weight: 400;\">Inndataene dine g\u00e5r inn, og resultatet kommer ut - og selv AI-utviklerne forst\u00e5r ikke helt hva som skjer inne i denne \"boksen\".\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Nevrale nettverk skaper sine egne interne representasjoner av informasjon n\u00e5r de tilordner inndata til utdata under datatreningen.\u00a0<\/span><span style=\"font-weight: 400;\">Byggesteinene i denne prosessen, kalt \"nevronaktiveringer\", representeres av numeriske verdier. <\/span><\/p>\n<p><span style=\"font-weight: 400;\">Hvert begrep er fordelt p\u00e5 flere nevroner, og hver nevron bidrar til \u00e5 representere flere begreper, noe som gj\u00f8r det vanskelig \u00e5 tilordne begreper direkte til enkeltnevroner.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Dette er i stor grad analogt med menneskehjernen. P\u00e5 samme m\u00e5te som hjernen v\u00e5r behandler sanseinntrykk og genererer tanker, atferd og minner, er de milliarder, til og med billioner, av prosesser som ligger bak disse funksjonene, stort sett ukjente for vitenskapen.<\/span><\/p>\n<p><a href=\"https:\/\/www.anthropic.com\/research\/mapping-mind-language-model\"><span style=\"font-weight: 400;\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\">Anthropic<\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/span>'s studie<\/span><\/a><span style=\"font-weight: 400;\"> fors\u00f8ker \u00e5 se inn i AIs svarte boks med en teknikk som kalles \"ordbokl\u00e6ring\".\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Dette inneb\u00e6rer \u00e5 dekomponere komplekse m\u00f8nstre i en AI-modell til line\u00e6re byggesteiner eller \"atomer\" som gir intuitiv mening for mennesker.<\/span><\/p>\n<h2>Kartlegging av LLM-er med ordbokl\u00e6ring<\/h2>\n<p><span style=\"font-weight: 400;\">I oktober 2023, <span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\">Anthropic<\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/span> brukte denne metoden p\u00e5 en liten \"leke\"-spr\u00e5kmodell og fant sammenhengende trekk som tilsvarte begreper som store bokstaver, DNA-sekvenser, etternavn i sitater, matematiske substantiver eller funksjonsargumenter i Python-kode.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Denne siste studien skalerer opp teknikken til \u00e5 fungere for dagens st\u00f8rre AI-spr\u00e5kmodeller, i dette tilfellet, <span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\">Anthropic<\/span><\/span><\/span><\/span>'s <span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\">Claude<\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/span> 3 Sonett.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Her er en trinnvis gjennomgang av hvordan studien fungerte:<\/span><\/p>\n<h4><b>Identifisere m\u00f8nstre med ordbokl\u00e6ring<\/b><\/h4>\n<p><span style=\"font-weight: 400;\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\">Anthropic<\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/span> brukte ordbokl\u00e6ring til \u00e5 analysere nevronaktiveringer i ulike kontekster og identifisere felles m\u00f8nstre. <\/span><\/p>\n<p><span style=\"font-weight: 400;\">Ordbokl\u00e6ring grupperer disse aktiveringene i et mindre sett med meningsfulle \"funksjoner\", som representerer begreper p\u00e5 h\u00f8yere niv\u00e5 som modellen har l\u00e6rt. <\/span><\/p>\n<p><span style=\"font-weight: 400;\">Ved \u00e5 identifisere disse funksjonene kan forskerne bedre forst\u00e5 hvordan modellen behandler og representerer informasjon.<\/span><\/p>\n<h4><b>Trekker ut funksjoner fra det midterste laget<\/b><\/h4>\n<p><span style=\"font-weight: 400;\">Forskerne fokuserte p\u00e5 det midterste laget av <span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\">Claude<\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/span> 3.0 Sonnet, som fungerer som et kritisk punkt i modellens prosesseringspipeline. <\/span><\/p>\n<p><span style=\"font-weight: 400;\">Ved \u00e5 bruke ordbokl\u00e6ring p\u00e5 dette laget ekstraheres millioner av funksjoner som fanger opp modellens interne representasjoner og innl\u00e6rte begreper p\u00e5 dette stadiet. <\/span><\/p>\n<p><span style=\"font-weight: 400;\">Ved \u00e5 trekke ut funksjoner fra det midterste laget kan forskerne unders\u00f8ke modellens forst\u00e5else av informasjon <em>etter<\/em> den har behandlet inndataene <em>f\u00f8r<\/em> som genererer det endelige resultatet.<\/span><\/p>\n<h4><b>Oppdage ulike og abstrakte begreper<\/b><\/h4>\n<p><span style=\"font-weight: 400;\">De ekstraherte funksjonene avsl\u00f8rte et ekspansivt spekter av begreper l\u00e6rt av <span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\">Claude<\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/span>fra konkrete enheter som byer og mennesker til abstrakte begreper knyttet til vitenskapelige felt og programmeringssyntaks. <\/span><\/p>\n<p><span style=\"font-weight: 400;\">Interessant nok viste det seg at funksjonene var multimodale og reagerte p\u00e5 b\u00e5de tekstlig og visuell input, noe som tyder p\u00e5 at modellen kan l\u00e6re og representere begreper p\u00e5 tvers av ulike modaliteter. <\/span><\/p>\n<p><span style=\"font-weight: 400;\">I tillegg tyder de flerspr\u00e5klige funksjonene p\u00e5 at modellen kan forst\u00e5 begreper som er uttrykt p\u00e5 ulike spr\u00e5k.<\/span><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"598\" class=\"wp-image-12521 size-large\" src=\"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/05\/c896a301ad3d1ef4237cb05b68d78b467c444097-2200x1284-1-1024x598.webp\" alt=\"&lt;span class=\" srcset=\"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/05\/c896a301ad3d1ef4237cb05b68d78b467c444097-2200x1284-1-1024x598.webp 1024w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/05\/c896a301ad3d1ef4237cb05b68d78b467c444097-2200x1284-1-300x175.webp 300w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/05\/c896a301ad3d1ef4237cb05b68d78b467c444097-2200x1284-1-768x448.webp 768w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/05\/c896a301ad3d1ef4237cb05b68d78b467c444097-2200x1284-1-1536x896.webp 1536w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/05\/c896a301ad3d1ef4237cb05b68d78b467c444097-2200x1284-1-2048x1195.webp 2048w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/05\/c896a301ad3d1ef4237cb05b68d78b467c444097-2200x1284-1-18x12.webp 18w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/05\/c896a301ad3d1ef4237cb05b68d78b467c444097-2200x1284-1-60x35.webp 60w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/p>\n<h3><b>Analysere organiseringen av konsepter<\/b><\/h3>\n<p><span style=\"font-weight: 400;\">For \u00e5 forst\u00e5 hvordan modellen organiserer og relaterer ulike konsepter, analyserte forskerne likheten mellom funksjonene basert p\u00e5 aktiveringsm\u00f8nstrene deres. <\/span><\/p>\n<p><span style=\"font-weight: 400;\">De oppdaget at funksjoner som representerte relaterte konsepter, hadde en tendens til \u00e5 gruppere seg sammen. For eksempel viste det seg at funksjoner knyttet til byer eller vitenskapelige disipliner hadde st\u00f8rre likhet med hverandre enn med funksjoner som representerte urelaterte konsepter. <\/span><\/p>\n<p><span style=\"font-weight: 400;\">Dette tyder p\u00e5 at modellens interne organisering av begreper til en viss grad samsvarer med menneskelige intuisjoner om konseptuelle relasjoner.<\/span><\/p>\n<figure id=\"attachment_12520\" aria-describedby=\"caption-attachment-12520\" style=\"width: 1024px\" class=\"wp-caption aligncenter\"><img loading=\"lazy\" decoding=\"async\" class=\"wp-image-12520 size-large\" src=\"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/05\/fe4d42c004bf43efda0f5921adfedd2f8f42e417-2200x2140-1-1024x996.webp\" alt=\"antropisk\" width=\"1024\" height=\"996\" srcset=\"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/05\/fe4d42c004bf43efda0f5921adfedd2f8f42e417-2200x2140-1-1024x996.webp 1024w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/05\/fe4d42c004bf43efda0f5921adfedd2f8f42e417-2200x2140-1-300x292.webp 300w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/05\/fe4d42c004bf43efda0f5921adfedd2f8f42e417-2200x2140-1-768x747.webp 768w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/05\/fe4d42c004bf43efda0f5921adfedd2f8f42e417-2200x2140-1-1536x1494.webp 1536w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/05\/fe4d42c004bf43efda0f5921adfedd2f8f42e417-2200x2140-1-2048x1992.webp 2048w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/05\/fe4d42c004bf43efda0f5921adfedd2f8f42e417-2200x2140-1-12x12.webp 12w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/05\/fe4d42c004bf43efda0f5921adfedd2f8f42e417-2200x2140-1-60x58.webp 60w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/05\/fe4d42c004bf43efda0f5921adfedd2f8f42e417-2200x2140-1-24x24.webp 24w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/05\/fe4d42c004bf43efda0f5921adfedd2f8f42e417-2200x2140-1-48x48.webp 48w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><figcaption id=\"caption-attachment-12520\" class=\"wp-caption-text\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\">Anthropic<\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/span> klarte \u00e5 kartlegge abstrakte begreper som \"indre konflikt\". Kilde: <a href=\"https:\/\/www.anthropic.com\/research\/mapping-mind-language-model\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\">Anthropic<\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/a>.<\/figcaption><\/figure>\n<h4><b>Verifisering av funksjonene<\/b><\/h4>\n<p><span style=\"font-weight: 400;\">For \u00e5 bekrefte at de identifiserte funksjonene p\u00e5virker modellens atferd og resultater direkte, utf\u00f8rte forskerne eksperimenter med \"funksjonsstyring\". <\/span><\/p>\n<p><span style=\"font-weight: 400;\">Dette innebar at man selektivt forsterket eller undertrykte aktiveringen av bestemte funksjoner under modellens prosessering, og observerte hvordan dette p\u00e5virket responsene. <\/span><\/p>\n<p><span style=\"font-weight: 400;\">Ved \u00e5 manipulere individuelle funksjoner kunne forskerne etablere en direkte kobling mellom individuelle funksjoner og modellens atferd. For eksempel f\u00f8rte en forsterkning av en funksjon knyttet til en bestemt by til at modellen genererte bybaserte resultater, selv i irrelevante sammenhenger.<\/span><\/p>\n<p>Les hele studien <a href=\"https:\/\/transformer-circuits.pub\/2024\/scaling-monosemanticity\/index.html\" target=\"_blank\" rel=\"noopener\">her<\/a>.<\/p>\n<p><iframe loading=\"lazy\" title=\"Ordbokl\u00e6ring p\u00e5 Claude 3 Sonnet\" width=\"1080\" height=\"608\" src=\"https:\/\/www.youtube.com\/embed\/CJIbCV92d88?feature=oembed\" frameborder=\"0\" allow=\"accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share\" referrerpolicy=\"strict-origin-when-cross-origin\" allowfullscreen><\/iframe><\/p>\n<h2>Hvorfor tolkbarhet er avgj\u00f8rende for AI-sikkerhet<\/h2>\n<p><span style=\"font-weight: 400;\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\">Anthropic<\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/span>'s forskning er grunnleggende relevant for AI-tolkbarhet og, i forlengelsen av dette, sikkerhet.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Ved \u00e5 forst\u00e5 hvordan LLM-er behandler og representerer informasjon, kan forskere bedre forst\u00e5 og redusere risiko. Det <\/span><span style=\"font-weight: 400;\">legger grunnlaget for \u00e5 utvikle mer transparente og forklarlige AI-systemer.\u00a0<\/span><\/p>\n<p>Som <span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\">Anthropic<\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/span> \"Vi h\u00e5per at vi og andre kan bruke disse oppdagelsene til \u00e5 gj\u00f8re modeller tryggere. Det kan for eksempel v\u00e6re mulig \u00e5 bruke teknikkene som er beskrevet her, til \u00e5 overv\u00e5ke AI-systemer for farlig atferd (for eksempel \u00e5 lure brukeren), til \u00e5 styre dem mot \u00f8nskelige resultater (debiasing), eller til \u00e5 fjerne visse farlige emner helt.\"<\/p>\n<p><span style=\"font-weight: 400;\">\u00c5 f\u00e5 en bedre forst\u00e5else av atferden til kunstig intelligens blir stadig viktigere etter hvert som de blir allestedsn\u00e6rv\u00e6rende i kritiske beslutningsprosesser p\u00e5 omr\u00e5der som helse, finans og strafferett. Det bidrar ogs\u00e5 til \u00e5 avdekke \u00e5rsaken til <a href=\"https:\/\/dailyai.com\/nb\/2023\/06\/anthropic-releases-paper-highlighting-the-potential-bias-of-large-language-models\/\">skjevhet<\/a>, hallusinasjoner og annen u\u00f8nsket eller uforutsigbar atferd.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">For eksempel kan en <\/span><a href=\"https:\/\/dailyai.com\/nb\/2024\/01\/researchers-aim-to-determine-how-black-box-drug-discovery-models-work\/\"><span style=\"font-weight: 400;\">fersk unders\u00f8kelse<\/span><\/a><span style=\"font-weight: 400;\"> fra Universitetet i Bonn avdekket hvordan grafnevrale nettverk (GNN) som brukes til legemiddeloppdagelse, i stor grad baserer seg p\u00e5 \u00e5 gjenkalle likheter fra treningsdata i stedet for \u00e5 l\u00e6re seg nye, komplekse kjemiske interaksjoner. <\/span><\/p>\n<p><span style=\"font-weight: 400;\">Dette gj\u00f8r det vanskelig \u00e5 forst\u00e5 n\u00f8yaktig hvordan disse modellene finner nye interessante forbindelser.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">I fjor ble <\/span><a href=\"https:\/\/dailyai.com\/nb\/2023\/09\/the-uk-government-wants-to-see-inside-ais-black-box\/\"><span style=\"font-weight: 400;\">Den britiske regjeringen forhandlet<\/span><\/a><span style=\"font-weight: 400;\"> med store teknologigiganter som <span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\">OpenAI<\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/span> og <span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\">DeepMind<\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/span>som \u00f8nsker innsyn i AI-systemenes interne beslutningsprosesser.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Regulering som <a href=\"https:\/\/dailyai.com\/nb\/2024\/03\/the-eu-ai-act-passed-in-a-landslide-and-will-come-into-force-this-year\/\">EUs lov om kunstig intelligens<\/a> vil presse AI-selskapene til \u00e5 v\u00e6re mer \u00e5pne, selv om forretningshemmeligheter ser ut til \u00e5 forbli under l\u00e5s og sl\u00e5.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\">Anthropic<\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/span>'s forskning gir et glimt av hva som befinner seg inne i boksen ved \u00e5 \"kartlegge\" informasjon p\u00e5 tvers av modellen.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Sannheten er imidlertid at disse modellene er s\u00e5 omfattende at de ved <span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\"><span class=\"noTranslate\" data-no-translation=\"\">Anthropic<\/span><\/span><\/span><\/span><\/span><\/span><\/span><\/span>\"Vi tror det er ganske sannsynlig at vi mangler mange st\u00f8rrelsesordener, og at hvis vi \u00f8nsket \u00e5 f\u00e5 med alle funksjonene - i alle lag! - ville vi m\u00e5tte bruke mye mer databehandling enn det som trengs for \u00e5 trene opp de underliggende modellene.\"<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Det er et interessant poeng - \u00e5 reversere en modell er mer komplisert enn \u00e5 konstruere modellen i utgangspunktet.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Det minner om enormt kostbare nevrovitenskapelige prosjekter som <a href=\"https:\/\/dailyai.com\/nb\/2023\/08\/exploring-impact-of-the-human-brain-project-hbp-on-science-and-ai\/\">Human Brain Project (HBP)<\/a>som brukte milliarder p\u00e5 \u00e5 kartlegge v\u00e5r egen menneskelige hjerne, men som til slutt mislyktes.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Undervurder aldri hvor mye som ligger inne i den svarte boksen.\u00a0<\/span><\/p>","protected":false},"excerpt":{"rendered":"<p>Forskere fra Anthropic har lykkes med \u00e5 identifisere millioner av begreper i Claude Sonnet, en av deres avanserte LLM-er. AI-modeller anses ofte som svarte bokser, noe som betyr at du ikke kan \"se\" inn i dem for \u00e5 forst\u00e5 n\u00f8yaktig hvordan de fungerer.  N\u00e5r du gir en LLM en input, genererer den et svar, men det er uklart hva som ligger bak valgene den tar. Inndataene dine g\u00e5r inn, og utdataene kommer ut - og selv AI-utviklerne forst\u00e5r ikke helt hva som skjer inne i denne \"boksen\".  Nevrale nettverk skaper sine egne interne representasjoner av informasjon n\u00e5r de kartlegger input til output under datatrening. Byggesteinene i dette<\/p>","protected":false},"author":2,"featured_media":12522,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[84],"tags":[148,105],"class_list":["post-12519","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-industry","tag-anthropic","tag-machine-learning"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v27.4 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>Decoding the AI mind: Anthropic researchers peer inside the &quot;black box&quot; | DailyAI<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/dailyai.com\/nb\/2024\/05\/decoding-the-ai-mind-anthropic-researchers-peer-inside-the-black-box\/\" \/>\n<meta property=\"og:locale\" content=\"nb_NO\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Decoding the AI mind: Anthropic researchers peer inside the &quot;black box&quot; | DailyAI\" \/>\n<meta property=\"og:description\" content=\"Anthropic researchers successfully identified millions of concepts within Claude Sonnet, one of their advanced LLMs. AI models are often considered black boxes, meaning you can&#8217;t &#8216;see&#8217; inside them to understand exactly how they work.\u00a0 When you provide an LLM with an input, it generates a response, but the reasoning behind its choices isn\u2019t clear. Your input goes in, and the output comes out \u2013 and even the AI developers themselves don&#8217;t truly understand what happens inside that &#8216;box.&#8217;\u00a0 Neural networks create their own internal representations of information when they map inputs to outputs during data training.\u00a0The building blocks of this\" \/>\n<meta property=\"og:url\" content=\"https:\/\/dailyai.com\/nb\/2024\/05\/decoding-the-ai-mind-anthropic-researchers-peer-inside-the-black-box\/\" \/>\n<meta property=\"og:site_name\" content=\"DailyAI\" \/>\n<meta property=\"article:published_time\" content=\"2024-05-22T18:16:07+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2024-05-24T10:16:44+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/05\/DALL\u00b7E-2024-05-22-19.15.50-Anthropic-researchers-studying-artificial-intelligence-inside-a-high-tech-laboratory-with-a-prominent-black-box-in-the-center.-The-scene-features-adva.webp\" \/>\n\t<meta property=\"og:image:width\" content=\"1792\" \/>\n\t<meta property=\"og:image:height\" content=\"1024\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/webp\" \/>\n<meta name=\"author\" content=\"Sam Jeans\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@DailyAIOfficial\" \/>\n<meta name=\"twitter:site\" content=\"@DailyAIOfficial\" \/>\n<meta name=\"twitter:label1\" content=\"Skrevet av\" \/>\n\t<meta name=\"twitter:data1\" content=\"Sam Jeans\" \/>\n\t<meta name=\"twitter:label2\" content=\"Ansl. lesetid\" \/>\n\t<meta name=\"twitter:data2\" content=\"6 minutter\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"NewsArticle\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/05\\\/decoding-the-ai-mind-anthropic-researchers-peer-inside-the-black-box\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/05\\\/decoding-the-ai-mind-anthropic-researchers-peer-inside-the-black-box\\\/\"},\"author\":{\"name\":\"Sam Jeans\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/person\\\/711e81f945549438e8bbc579efdeb3c9\"},\"headline\":\"Decoding the AI mind: Anthropic researchers peer inside the &#8220;black box&#8221;\",\"datePublished\":\"2024-05-22T18:16:07+00:00\",\"dateModified\":\"2024-05-24T10:16:44+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/05\\\/decoding-the-ai-mind-anthropic-researchers-peer-inside-the-black-box\\\/\"},\"wordCount\":1075,\"publisher\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/05\\\/decoding-the-ai-mind-anthropic-researchers-peer-inside-the-black-box\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2024\\\/05\\\/DALL\u00b7E-2024-05-22-19.15.50-Anthropic-researchers-studying-artificial-intelligence-inside-a-high-tech-laboratory-with-a-prominent-black-box-in-the-center.-The-scene-features-adva.webp\",\"keywords\":[\"Anthropic\",\"machine learning\"],\"articleSection\":[\"Industry\"],\"inLanguage\":\"nb-NO\"},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/05\\\/decoding-the-ai-mind-anthropic-researchers-peer-inside-the-black-box\\\/\",\"url\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/05\\\/decoding-the-ai-mind-anthropic-researchers-peer-inside-the-black-box\\\/\",\"name\":\"Decoding the AI mind: Anthropic researchers peer inside the \\\"black box\\\" | DailyAI\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/05\\\/decoding-the-ai-mind-anthropic-researchers-peer-inside-the-black-box\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/05\\\/decoding-the-ai-mind-anthropic-researchers-peer-inside-the-black-box\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2024\\\/05\\\/DALL\u00b7E-2024-05-22-19.15.50-Anthropic-researchers-studying-artificial-intelligence-inside-a-high-tech-laboratory-with-a-prominent-black-box-in-the-center.-The-scene-features-adva.webp\",\"datePublished\":\"2024-05-22T18:16:07+00:00\",\"dateModified\":\"2024-05-24T10:16:44+00:00\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/05\\\/decoding-the-ai-mind-anthropic-researchers-peer-inside-the-black-box\\\/#breadcrumb\"},\"inLanguage\":\"nb-NO\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/dailyai.com\\\/2024\\\/05\\\/decoding-the-ai-mind-anthropic-researchers-peer-inside-the-black-box\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"nb-NO\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/05\\\/decoding-the-ai-mind-anthropic-researchers-peer-inside-the-black-box\\\/#primaryimage\",\"url\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2024\\\/05\\\/DALL\u00b7E-2024-05-22-19.15.50-Anthropic-researchers-studying-artificial-intelligence-inside-a-high-tech-laboratory-with-a-prominent-black-box-in-the-center.-The-scene-features-adva.webp\",\"contentUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2024\\\/05\\\/DALL\u00b7E-2024-05-22-19.15.50-Anthropic-researchers-studying-artificial-intelligence-inside-a-high-tech-laboratory-with-a-prominent-black-box-in-the-center.-The-scene-features-adva.webp\",\"width\":1792,\"height\":1024},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/05\\\/decoding-the-ai-mind-anthropic-researchers-peer-inside-the-black-box\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/dailyai.com\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Decoding the AI mind: Anthropic researchers peer inside the &#8220;black box&#8221;\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#website\",\"url\":\"https:\\\/\\\/dailyai.com\\\/\",\"name\":\"DailyAI\",\"description\":\"Your Daily Dose of AI News\",\"publisher\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/dailyai.com\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"nb-NO\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\",\"name\":\"DailyAI\",\"url\":\"https:\\\/\\\/dailyai.com\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"nb-NO\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/logo\\\/image\\\/\",\"url\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/06\\\/Daily-Ai_TL_colour.png\",\"contentUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/06\\\/Daily-Ai_TL_colour.png\",\"width\":4501,\"height\":934,\"caption\":\"DailyAI\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/logo\\\/image\\\/\"},\"sameAs\":[\"https:\\\/\\\/x.com\\\/DailyAIOfficial\",\"https:\\\/\\\/www.linkedin.com\\\/company\\\/dailyaiofficial\\\/\",\"https:\\\/\\\/www.youtube.com\\\/@DailyAIOfficial\"]},{\"@type\":\"Person\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/person\\\/711e81f945549438e8bbc579efdeb3c9\",\"name\":\"Sam Jeans\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"nb-NO\",\"@id\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/a24a4a8f8e2a1a275b7491dc9c9f032c401eabf23c3206da4628dc84b6dac5c8?s=96&d=robohash&r=g\",\"url\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/a24a4a8f8e2a1a275b7491dc9c9f032c401eabf23c3206da4628dc84b6dac5c8?s=96&d=robohash&r=g\",\"contentUrl\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/a24a4a8f8e2a1a275b7491dc9c9f032c401eabf23c3206da4628dc84b6dac5c8?s=96&d=robohash&r=g\",\"caption\":\"Sam Jeans\"},\"description\":\"Sam is a science and technology writer who has worked in various AI startups. When he\u2019s not writing, he can be found reading medical journals or digging through boxes of vinyl records.\",\"sameAs\":[\"https:\\\/\\\/www.linkedin.com\\\/in\\\/sam-jeans-6746b9142\\\/\"],\"url\":\"https:\\\/\\\/dailyai.com\\\/nb\\\/author\\\/samjeans\\\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Avkoding av AI-sinnet: Antropiske forskere kikker inn i den \"svarte boksen\" | DailyAI","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/dailyai.com\/nb\/2024\/05\/decoding-the-ai-mind-anthropic-researchers-peer-inside-the-black-box\/","og_locale":"nb_NO","og_type":"article","og_title":"Decoding the AI mind: Anthropic researchers peer inside the \"black box\" | DailyAI","og_description":"Anthropic researchers successfully identified millions of concepts within Claude Sonnet, one of their advanced LLMs. AI models are often considered black boxes, meaning you can&#8217;t &#8216;see&#8217; inside them to understand exactly how they work.\u00a0 When you provide an LLM with an input, it generates a response, but the reasoning behind its choices isn\u2019t clear. Your input goes in, and the output comes out \u2013 and even the AI developers themselves don&#8217;t truly understand what happens inside that &#8216;box.&#8217;\u00a0 Neural networks create their own internal representations of information when they map inputs to outputs during data training.\u00a0The building blocks of this","og_url":"https:\/\/dailyai.com\/nb\/2024\/05\/decoding-the-ai-mind-anthropic-researchers-peer-inside-the-black-box\/","og_site_name":"DailyAI","article_published_time":"2024-05-22T18:16:07+00:00","article_modified_time":"2024-05-24T10:16:44+00:00","og_image":[{"width":1792,"height":1024,"url":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/05\/DALL\u00b7E-2024-05-22-19.15.50-Anthropic-researchers-studying-artificial-intelligence-inside-a-high-tech-laboratory-with-a-prominent-black-box-in-the-center.-The-scene-features-adva.webp","type":"image\/webp"}],"author":"Sam Jeans","twitter_card":"summary_large_image","twitter_creator":"@DailyAIOfficial","twitter_site":"@DailyAIOfficial","twitter_misc":{"Skrevet av":"Sam Jeans","Ansl. lesetid":"6 minutter"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"NewsArticle","@id":"https:\/\/dailyai.com\/2024\/05\/decoding-the-ai-mind-anthropic-researchers-peer-inside-the-black-box\/#article","isPartOf":{"@id":"https:\/\/dailyai.com\/2024\/05\/decoding-the-ai-mind-anthropic-researchers-peer-inside-the-black-box\/"},"author":{"name":"Sam Jeans","@id":"https:\/\/dailyai.com\/#\/schema\/person\/711e81f945549438e8bbc579efdeb3c9"},"headline":"Decoding the AI mind: Anthropic researchers peer inside the &#8220;black box&#8221;","datePublished":"2024-05-22T18:16:07+00:00","dateModified":"2024-05-24T10:16:44+00:00","mainEntityOfPage":{"@id":"https:\/\/dailyai.com\/2024\/05\/decoding-the-ai-mind-anthropic-researchers-peer-inside-the-black-box\/"},"wordCount":1075,"publisher":{"@id":"https:\/\/dailyai.com\/#organization"},"image":{"@id":"https:\/\/dailyai.com\/2024\/05\/decoding-the-ai-mind-anthropic-researchers-peer-inside-the-black-box\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/05\/DALL\u00b7E-2024-05-22-19.15.50-Anthropic-researchers-studying-artificial-intelligence-inside-a-high-tech-laboratory-with-a-prominent-black-box-in-the-center.-The-scene-features-adva.webp","keywords":["Anthropic","machine learning"],"articleSection":["Industry"],"inLanguage":"nb-NO"},{"@type":"WebPage","@id":"https:\/\/dailyai.com\/2024\/05\/decoding-the-ai-mind-anthropic-researchers-peer-inside-the-black-box\/","url":"https:\/\/dailyai.com\/2024\/05\/decoding-the-ai-mind-anthropic-researchers-peer-inside-the-black-box\/","name":"Avkoding av AI-sinnet: Antropiske forskere kikker inn i den \"svarte boksen\" | DailyAI","isPartOf":{"@id":"https:\/\/dailyai.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/dailyai.com\/2024\/05\/decoding-the-ai-mind-anthropic-researchers-peer-inside-the-black-box\/#primaryimage"},"image":{"@id":"https:\/\/dailyai.com\/2024\/05\/decoding-the-ai-mind-anthropic-researchers-peer-inside-the-black-box\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/05\/DALL\u00b7E-2024-05-22-19.15.50-Anthropic-researchers-studying-artificial-intelligence-inside-a-high-tech-laboratory-with-a-prominent-black-box-in-the-center.-The-scene-features-adva.webp","datePublished":"2024-05-22T18:16:07+00:00","dateModified":"2024-05-24T10:16:44+00:00","breadcrumb":{"@id":"https:\/\/dailyai.com\/2024\/05\/decoding-the-ai-mind-anthropic-researchers-peer-inside-the-black-box\/#breadcrumb"},"inLanguage":"nb-NO","potentialAction":[{"@type":"ReadAction","target":["https:\/\/dailyai.com\/2024\/05\/decoding-the-ai-mind-anthropic-researchers-peer-inside-the-black-box\/"]}]},{"@type":"ImageObject","inLanguage":"nb-NO","@id":"https:\/\/dailyai.com\/2024\/05\/decoding-the-ai-mind-anthropic-researchers-peer-inside-the-black-box\/#primaryimage","url":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/05\/DALL\u00b7E-2024-05-22-19.15.50-Anthropic-researchers-studying-artificial-intelligence-inside-a-high-tech-laboratory-with-a-prominent-black-box-in-the-center.-The-scene-features-adva.webp","contentUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/05\/DALL\u00b7E-2024-05-22-19.15.50-Anthropic-researchers-studying-artificial-intelligence-inside-a-high-tech-laboratory-with-a-prominent-black-box-in-the-center.-The-scene-features-adva.webp","width":1792,"height":1024},{"@type":"BreadcrumbList","@id":"https:\/\/dailyai.com\/2024\/05\/decoding-the-ai-mind-anthropic-researchers-peer-inside-the-black-box\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/dailyai.com\/"},{"@type":"ListItem","position":2,"name":"Decoding the AI mind: Anthropic researchers peer inside the &#8220;black box&#8221;"}]},{"@type":"WebSite","@id":"https:\/\/dailyai.com\/#website","url":"https:\/\/dailyai.com\/","name":"DagligAI","description":"Din daglige dose med AI-nyheter","publisher":{"@id":"https:\/\/dailyai.com\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/dailyai.com\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"nb-NO"},{"@type":"Organization","@id":"https:\/\/dailyai.com\/#organization","name":"DagligAI","url":"https:\/\/dailyai.com\/","logo":{"@type":"ImageObject","inLanguage":"nb-NO","@id":"https:\/\/dailyai.com\/#\/schema\/logo\/image\/","url":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/06\/Daily-Ai_TL_colour.png","contentUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/06\/Daily-Ai_TL_colour.png","width":4501,"height":934,"caption":"DailyAI"},"image":{"@id":"https:\/\/dailyai.com\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/x.com\/DailyAIOfficial","https:\/\/www.linkedin.com\/company\/dailyaiofficial\/","https:\/\/www.youtube.com\/@DailyAIOfficial"]},{"@type":"Person","@id":"https:\/\/dailyai.com\/#\/schema\/person\/711e81f945549438e8bbc579efdeb3c9","name":"Sam Jeans","image":{"@type":"ImageObject","inLanguage":"nb-NO","@id":"https:\/\/secure.gravatar.com\/avatar\/a24a4a8f8e2a1a275b7491dc9c9f032c401eabf23c3206da4628dc84b6dac5c8?s=96&d=robohash&r=g","url":"https:\/\/secure.gravatar.com\/avatar\/a24a4a8f8e2a1a275b7491dc9c9f032c401eabf23c3206da4628dc84b6dac5c8?s=96&d=robohash&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/a24a4a8f8e2a1a275b7491dc9c9f032c401eabf23c3206da4628dc84b6dac5c8?s=96&d=robohash&r=g","caption":"Sam Jeans"},"description":"Sam er en vitenskaps- og teknologiskribent som har jobbet i ulike oppstartsbedrifter innen kunstig intelligens. N\u00e5r han ikke skriver, leser han medisinske tidsskrifter eller graver seg gjennom esker med vinylplater.","sameAs":["https:\/\/www.linkedin.com\/in\/sam-jeans-6746b9142\/"],"url":"https:\/\/dailyai.com\/nb\/author\/samjeans\/"}]}},"_links":{"self":[{"href":"https:\/\/dailyai.com\/nb\/wp-json\/wp\/v2\/posts\/12519","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/dailyai.com\/nb\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/dailyai.com\/nb\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/dailyai.com\/nb\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/dailyai.com\/nb\/wp-json\/wp\/v2\/comments?post=12519"}],"version-history":[{"count":11,"href":"https:\/\/dailyai.com\/nb\/wp-json\/wp\/v2\/posts\/12519\/revisions"}],"predecessor-version":[{"id":12573,"href":"https:\/\/dailyai.com\/nb\/wp-json\/wp\/v2\/posts\/12519\/revisions\/12573"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/dailyai.com\/nb\/wp-json\/wp\/v2\/media\/12522"}],"wp:attachment":[{"href":"https:\/\/dailyai.com\/nb\/wp-json\/wp\/v2\/media?parent=12519"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/dailyai.com\/nb\/wp-json\/wp\/v2\/categories?post=12519"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/dailyai.com\/nb\/wp-json\/wp\/v2\/tags?post=12519"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}