{"id":6203,"date":"2023-10-07T17:15:19","date_gmt":"2023-10-07T17:15:19","guid":{"rendered":"https:\/\/dailyai.com\/?p=6203"},"modified":"2023-10-07T22:49:19","modified_gmt":"2023-10-07T22:49:19","slug":"can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior","status":"publish","type":"post","link":"https:\/\/dailyai.com\/nb\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/","title":{"rendered":"Kan \"konstitusjonell AI\" l\u00f8se problemet med problematisk AI-atferd?"},"content":{"rendered":"

I takt med at AI-modeller blir en stadig st\u00f8rre del av hverdagen v\u00e5r, \u00f8ker bekymringene for begrensningene og p\u00e5liteligheten til de s\u00e5kalte \"beskyttelsesrammene\".<\/b><\/p>\n

Allestedsn\u00e6rv\u00e6rende AI-modeller som GPT-3.5\/4\/4V m.fl. har innebygde rekkverk og sikkerhetstiltak for \u00e5 forhindre at de produserer ulovlige, uetiske eller p\u00e5 annen m\u00e5te u\u00f8nskede resultater. <\/span><\/p>\n

Disse sikkerhetsfunksjonene er imidlertid langt fra ugjennomtrengelige, og flere modeller har vist seg \u00e5 kunne l\u00f8sne fra rekkverket - eller g\u00e5 av sporet, for \u00e5 si det s\u00e5nn. <\/span><\/p>\n

En del av problemet er at beskyttelsesrammene ikke holder tritt med modellenes kompleksitet og mangfold.\u00a0<\/span><\/p>\n

De siste ukene har OpenAI, som st\u00f8ttes av Microsoft, avsl\u00f8rt store forbedringer i ChatGPT, som gj\u00f8r det mulig \u00e5 samhandle kun ved hjelp av tale og svare p\u00e5 sp\u00f8rsm\u00e5l gjennom bilder og tekst. Denne multimodale versjonen av GPT-4, som er kompatibel med bilder, har f\u00e5tt navnet \"GPT-4V\".<\/span><\/p>\n

Parallelt med dette kunngjorde Meta lanseringen av en AI-assistent, flere kjendis-chatbot-personligheter<\/a> for WhatsApp- og Instagram-brukere, og en rekke andre lavm\u00e6lte AI-funksjoner som AI-klistremerker.\u00a0<\/span><\/p>\n

Folk manipulerte raskt Metas <\/span>AI-klistremerker for \u00e5 generere<\/span><\/a> komiske og sjokkerende tegneserielignende bilder, som Karl Marx naken eller Mario med automatgev\u00e6r.\u00a0<\/span><\/p>\n

I takt med at kappl\u00f8pet om \u00e5 kommersialisere kunstig intelligens intensiveres, blir sikkerhetstiltakene som skal kontrollere atferden til kunstig intelligens - og forhindre at den genererer skadelig innhold, feilinformasjon eller medvirker til ulovlige aktiviteter - stadig svakere.\u00a0<\/span><\/p>\n

Er konstitusjonell kunstig intelligens svaret?<\/span><\/h2>\n

For \u00e5 bekjempe dette fors\u00f8ker AI-utviklere \u00e5 skape \"AI-konstitusjoner\", et sett med grunnleggende prinsipper og verdier som AI-modeller m\u00e5 forholde seg til. Oppstarten Antropisk<\/a> var blant de f\u00f8rste som tok til orde for \"konstitusjonell AI\" i en <\/span>2022 papir<\/span><\/a>.<\/span><\/p>\n

Google DeepMind etablerte ogs\u00e5 konstitusjonelle regler for chatboten sin Spurv i 2022<\/a> \u00e5 f\u00f8re \"hjelpsomme, korrekte og ufarlige\" samtaler.\u00a0<\/span><\/p>\n

Anthropics AI-konstitusjoner henter prinsipper fra ulike kilder, blant annet FNs menneskerettighetserkl\u00e6ring og Apples brukervilk\u00e5r. Modellen er utstyrt med grunnleggende moralske prinsipper som styrer atferden nedenfra og opp, i stedet for \u00e5 p\u00e5legge oss retningslinjer ovenfra og ned.\u00a0<\/span><\/p>\n

I stedet for \u00e5 trene opp kunstig intelligens med utallige eksempler p\u00e5 hva som er rett og galt, bygger denne tiln\u00e6rmingen inn et sett med regler eller prinsipper - en \"grunnlov\" - som den kunstige intelligensen f\u00f8lger.<\/p>\n

F\u00f8rst blir den kunstige intelligensen introdusert for en situasjon, deretter blir den bedt om \u00e5 kritisere responsen, og til slutt finjusterer den atferden sin basert p\u00e5 den reviderte l\u00f8sningen.<\/p>\n

Deretter dykker systemet ned i forsterkningsl\u00e6ringsfasen. Her m\u00e5ler det kvaliteten p\u00e5 sine egne svar, og skiller ut de beste. Over tid forbedrer denne egenvurderingen atferden.<\/p>\n

Det nye er at den kunstige intelligensen bruker sin egen tilbakemeldingssl\u00f8yfe til \u00e5 fastsette bel\u00f8nningen ved hjelp av en metode som kalles \"RL from AI Feedback\" (RLAIF). N\u00e5r AI-en blir konfrontert med potensielt skadelige eller villedende foresp\u00f8rsler, unng\u00e5r den ikke bare \u00e5 svare eller nekte. I stedet tar den direkte tak i saken og forklarer hvorfor en slik foresp\u00f8rsel kan v\u00e6re problematisk.<\/p>\n

Det er et skritt fremover i arbeidet med \u00e5 skape maskiner som ikke bare beregner, men som ogs\u00e5 \"tenker\" p\u00e5 en strukturert m\u00e5te.<\/p>\n

Dario Amodei, administrerende direkt\u00f8r og medgrunnlegger av Anthropic, understreket utfordringen med \u00e5 forst\u00e5 hvordan AI-modeller fungerer. Han foresl\u00e5r at en grunnlov vil gj\u00f8re reglene transparente og eksplisitte, slik at alle brukere vet hva de kan forvente.\u00a0<\/span><\/p>\n

Det er ogs\u00e5 viktig at modellen kan holdes ansvarlig hvis den ikke f\u00f8lger de skisserte prinsippene.<\/span><\/p>\n

Til tross for denne innsatsen er AI-konstitusjonene ikke uten egne feil, og modeller fra utviklere som Anthropic har vist seg \u00e5 v\u00e6re s\u00e5rbare for <\/span>jailbreaks<\/span><\/a> som s\u00e5 mange andre.\u00a0<\/span><\/p>\n

Det finnes ingen universelt aksepterte m\u00e5ter \u00e5 trene opp trygge og etiske AI-modeller p\u00e5<\/span><\/h2>\n

Historisk sett har AI-modeller blitt raffinert ved hjelp av en metode som kalles reinforcement learning by human feedback (RLHF), der AI-responser kategoriseres som \"gode\" eller \"d\u00e5rlige\" av store team med menneskelige evaluatorer.\u00a0<\/span><\/p>\n

Selv om denne metoden til en viss grad er effektiv, har den blitt kritisert for sin mangel p\u00e5 n\u00f8yaktighet og spesifisitet. For \u00e5 sikre etikk og sikkerhet i forbindelse med kunstig intelligens utforsker selskapene n\u00e5 alternative l\u00f8sninger.<\/span><\/p>\n

OpenAI har for eksempel tatt i bruk \"red-teaming\"-metoden, der de ansetter eksperter fra ulike fagomr\u00e5der for \u00e5 teste og identifisere svakheter i modellene sine.<\/span><\/p>\n

OpenAIs system fungerer i iterasjoner: AI-modellen produserer resultater, menneskelige anmeldere vurderer og korrigerer disse resultatene basert p\u00e5 spesifikke retningslinjer, og modellen l\u00e6rer av denne tilbakemeldingen. Oppl\u00e6ringsdataene fra disse anmelderne er avgj\u00f8rende for modellens etiske kalibrering.<\/p>\n

ChatGPT velger ofte et konservativt svar n\u00e5r den blir konfrontert med kontroversielle eller sensitive emner, og unng\u00e5r noen ganger et direkte svar. Dette st\u00e5r i kontrast til konstitusjonell AI, der modellen b\u00f8r tydeliggj\u00f8re sine reservasjoner n\u00e5r den blir stilt overfor potensielt skadelige sp\u00f8rsm\u00e5l, og aktivt demonstrere resonnementer basert p\u00e5 sine grunnleggende regler.<\/p>\n

Mens ChatGPT i stor grad baserer seg p\u00e5 menneskelig tilbakemelding for sin etiske orientering, bruker den konstitusjonelle AI-en et regelbasert rammeverk med mekanismer for selvransakelse og vekt p\u00e5 transparente resonnementer.<\/p>\n

Til syvende og sist finnes det sannsynligvis ingen universell tiln\u00e6rming til \u00e5 utvikle \"trygg\" AI - og noen, som Elon Musk, kritiserer forestillingen om \"v\u00e5ken\" AI. Studier har vist at<\/a> at selv konstitusjonelle AI-er kan brytes ned og manipuleres til \u00e5 oppf\u00f8re seg uforutsigbart.\u00a0<\/span><\/p>\n

Rebecca Johnson, som forsker p\u00e5 AI-etikk ved University of Sydney, p\u00e5pekte at AI-ingeni\u00f8rer og dataforskere ofte tiln\u00e6rmer seg problemer med sikte p\u00e5 \u00e5 finne endelige l\u00f8sninger, noe som kanskje ikke alltid tar hensyn til kompleksiteten i menneskets natur.\u00a0<\/span><\/p>\n

\"Vi m\u00e5 begynne \u00e5 behandle generativ AI som en forlengelse av mennesket, de er bare et annet aspekt av menneskeheten\", sier hun.\u00a0<\/span><\/p>\n

Omfattende kontroll av AI som et slags enkelt teknisk system vil bare bli <\/span>vanskeligere etter hvert som den utvikler seg<\/span><\/a>Det samme kan sies om biologiske organismer som oss selv. <\/span><\/p>\n

Uenighet, provosert eller ikke, er kanskje uunng\u00e5elig. <\/span><\/p>","protected":false},"excerpt":{"rendered":"

Etter hvert som AI-modeller blir en stadig st\u00f8rre del av hverdagen v\u00e5r, \u00f8ker bekymringene for begrensningene og p\u00e5liteligheten til de s\u00e5kalte \"beskyttelsesrammene\". Allestedsn\u00e6rv\u00e6rende AI-modeller som GPT-3.5\/4\/4V m.fl. har innebygde rekkverk og sikkerhetstiltak som skal hindre dem i \u00e5 produsere ulovlige, uetiske eller p\u00e5 annen m\u00e5te u\u00f8nskede resultater. Disse sikkerhetstiltakene er imidlertid langt fra ugjennomtrengelige, og modellene har vist seg \u00e5 kunne l\u00f8srive seg fra rekkverkene - eller g\u00e5 av sporet, for \u00e5 si det slik. En del av problemet er at rekkverkene ikke holder tritt med modellenes kompleksitet og mangfold. I l\u00f8pet av de siste ukene har OpenAI, som st\u00f8ttes av Microsoft, avsl\u00f8rt store forbedringer<\/p>","protected":false},"author":2,"featured_media":6204,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[84],"tags":[148,422,93],"class_list":["post-6203","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-industry","tag-anthropic","tag-constitutional-ai","tag-openai"],"yoast_head":"\nCan "constitutional AI" solve the issue of problematic AI behavior? | DailyAI<\/title>\n<meta name=\"description\" content=\"As AI models continue to embed themselves in our daily lives, concerns over the limitations and reliability of the so-called "guardrails" are mounting.\" \/>\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\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/\" \/>\n<meta property=\"og:locale\" content=\"nb_NO\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Can "constitutional AI" solve the issue of problematic AI behavior? | DailyAI\" \/>\n<meta property=\"og:description\" content=\"As AI models continue to embed themselves in our daily lives, concerns over the limitations and reliability of the so-called "guardrails" are mounting.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/dailyai.com\/nb\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/\" \/>\n<meta property=\"og:site_name\" content=\"DailyAI\" \/>\n<meta property=\"article:published_time\" content=\"2023-10-07T17:15:19+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2023-10-07T22:49:19+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/shutterstock_2276608417-1.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"1000\" \/>\n\t<meta property=\"og:image:height\" content=\"667\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\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=\"4 minutter\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"NewsArticle\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\\\/\"},\"author\":{\"name\":\"Sam Jeans\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/person\\\/711e81f945549438e8bbc579efdeb3c9\"},\"headline\":\"Can “constitutional AI” solve the issue of problematic AI behavior?\",\"datePublished\":\"2023-10-07T17:15:19+00:00\",\"dateModified\":\"2023-10-07T22:49:19+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\\\/\"},\"wordCount\":915,\"publisher\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/10\\\/shutterstock_2276608417-1.jpg\",\"keywords\":[\"Anthropic\",\"Constitutional AI\",\"OpenAI\"],\"articleSection\":[\"Industry\"],\"inLanguage\":\"nb-NO\"},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\\\/\",\"url\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\\\/\",\"name\":\"Can \\\"constitutional AI\\\" solve the issue of problematic AI behavior? | DailyAI\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/10\\\/shutterstock_2276608417-1.jpg\",\"datePublished\":\"2023-10-07T17:15:19+00:00\",\"dateModified\":\"2023-10-07T22:49:19+00:00\",\"description\":\"As AI models continue to embed themselves in our daily lives, concerns over the limitations and reliability of the so-called \\\"guardrails\\\" are mounting.\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\\\/#breadcrumb\"},\"inLanguage\":\"nb-NO\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"nb-NO\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\\\/#primaryimage\",\"url\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/10\\\/shutterstock_2276608417-1.jpg\",\"contentUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/10\\\/shutterstock_2276608417-1.jpg\",\"width\":1000,\"height\":667,\"caption\":\"Anthropic AI\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/dailyai.com\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Can “constitutional AI” solve the issue of problematic AI behavior?\"}]},{\"@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":"Kan \"konstitusjonell AI\" l\u00f8se problemet med problematisk AI-atferd? | DailyAI","description":"Etter hvert som AI-modeller blir en stadig st\u00f8rre del av hverdagen v\u00e5r, \u00f8ker bekymringene for begrensningene og p\u00e5liteligheten til de s\u00e5kalte \"rekkverkene\".","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\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/","og_locale":"nb_NO","og_type":"article","og_title":"Can \"constitutional AI\" solve the issue of problematic AI behavior? | DailyAI","og_description":"As AI models continue to embed themselves in our daily lives, concerns over the limitations and reliability of the so-called \"guardrails\" are mounting.","og_url":"https:\/\/dailyai.com\/nb\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/","og_site_name":"DailyAI","article_published_time":"2023-10-07T17:15:19+00:00","article_modified_time":"2023-10-07T22:49:19+00:00","og_image":[{"width":1000,"height":667,"url":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/shutterstock_2276608417-1.jpg","type":"image\/jpeg"}],"author":"Sam Jeans","twitter_card":"summary_large_image","twitter_creator":"@DailyAIOfficial","twitter_site":"@DailyAIOfficial","twitter_misc":{"Skrevet av":"Sam Jeans","Ansl. lesetid":"4 minutter"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"NewsArticle","@id":"https:\/\/dailyai.com\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/#article","isPartOf":{"@id":"https:\/\/dailyai.com\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/"},"author":{"name":"Sam Jeans","@id":"https:\/\/dailyai.com\/#\/schema\/person\/711e81f945549438e8bbc579efdeb3c9"},"headline":"Can “constitutional AI” solve the issue of problematic AI behavior?","datePublished":"2023-10-07T17:15:19+00:00","dateModified":"2023-10-07T22:49:19+00:00","mainEntityOfPage":{"@id":"https:\/\/dailyai.com\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/"},"wordCount":915,"publisher":{"@id":"https:\/\/dailyai.com\/#organization"},"image":{"@id":"https:\/\/dailyai.com\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/shutterstock_2276608417-1.jpg","keywords":["Anthropic","Constitutional AI","OpenAI"],"articleSection":["Industry"],"inLanguage":"nb-NO"},{"@type":"WebPage","@id":"https:\/\/dailyai.com\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/","url":"https:\/\/dailyai.com\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/","name":"Kan \"konstitusjonell AI\" l\u00f8se problemet med problematisk AI-atferd? | DailyAI","isPartOf":{"@id":"https:\/\/dailyai.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/dailyai.com\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/#primaryimage"},"image":{"@id":"https:\/\/dailyai.com\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/shutterstock_2276608417-1.jpg","datePublished":"2023-10-07T17:15:19+00:00","dateModified":"2023-10-07T22:49:19+00:00","description":"Etter hvert som AI-modeller blir en stadig st\u00f8rre del av hverdagen v\u00e5r, \u00f8ker bekymringene for begrensningene og p\u00e5liteligheten til de s\u00e5kalte \"rekkverkene\".","breadcrumb":{"@id":"https:\/\/dailyai.com\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/#breadcrumb"},"inLanguage":"nb-NO","potentialAction":[{"@type":"ReadAction","target":["https:\/\/dailyai.com\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/"]}]},{"@type":"ImageObject","inLanguage":"nb-NO","@id":"https:\/\/dailyai.com\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/#primaryimage","url":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/shutterstock_2276608417-1.jpg","contentUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/shutterstock_2276608417-1.jpg","width":1000,"height":667,"caption":"Anthropic AI"},{"@type":"BreadcrumbList","@id":"https:\/\/dailyai.com\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/dailyai.com\/"},{"@type":"ListItem","position":2,"name":"Can “constitutional AI” solve the issue of problematic AI behavior?"}]},{"@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\/6203","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=6203"}],"version-history":[{"count":12,"href":"https:\/\/dailyai.com\/nb\/wp-json\/wp\/v2\/posts\/6203\/revisions"}],"predecessor-version":[{"id":6216,"href":"https:\/\/dailyai.com\/nb\/wp-json\/wp\/v2\/posts\/6203\/revisions\/6216"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/dailyai.com\/nb\/wp-json\/wp\/v2\/media\/6204"}],"wp:attachment":[{"href":"https:\/\/dailyai.com\/nb\/wp-json\/wp\/v2\/media?parent=6203"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/dailyai.com\/nb\/wp-json\/wp\/v2\/categories?post=6203"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/dailyai.com\/nb\/wp-json\/wp\/v2\/tags?post=6203"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}