{"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\/sv\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/","title":{"rendered":"Kan \"konstitutionell AI\" l\u00f6sa fr\u00e5gan om problematiskt AI-beteende?"},"content":{"rendered":"

I takt med att AI-modellerna forts\u00e4tter att integreras i v\u00e5ra dagliga liv \u00f6kar oron \u00f6ver begr\u00e4nsningarna och tillf\u00f6rlitligheten hos deras s\u00e5 kallade \"skyddsr\u00e4cken\".<\/b><\/p>\n

Allest\u00e4des n\u00e4rvarande AI-modeller som GPT-3.5\/4\/4V et al. har inbyggda skyddsr\u00e4cken och s\u00e4kerhets\u00e5tg\u00e4rder f\u00f6r att f\u00f6rhindra att de producerar olagliga, oetiska eller p\u00e5 annat s\u00e4tt o\u00f6nskade resultat. <\/span><\/p>\n

Dessa s\u00e4kerhetsfunktioner \u00e4r dock l\u00e5ngt ifr\u00e5n ogenomtr\u00e4ngliga och det finns modeller som visar att de kan lossna fr\u00e5n sina skyddsr\u00e4cken - eller sp\u00e5ra ur, s\u00e5 att s\u00e4ga. <\/span><\/p>\n

En del av problemet \u00e4r att skyddsr\u00e4ckena inte h\u00e5ller j\u00e4mna steg med modellernas komplexitet och m\u00e5ngfald.\u00a0<\/span><\/p>\n

Under de senaste veckorna har OpenAI, med st\u00f6d av Microsoft, presenterat stora f\u00f6rb\u00e4ttringar av ChatGPT som g\u00f6r det m\u00f6jligt att interagera med enbart r\u00f6st och svara p\u00e5 fr\u00e5gor via bilder och text. Denna multimodala version av GPT-4 som kan hantera bilder har f\u00e5tt namnet \"GPT-4V\".<\/span><\/p>\n

Parallellt meddelade Meta utrullningen av en AI-assistent, flera k\u00e4ndis chatbot personligheter<\/a> f\u00f6r WhatsApp- och Instagram-anv\u00e4ndare, och en m\u00e4ngd andra l\u00e5gm\u00e4lda AI-funktioner som AI Stickers.\u00a0<\/span><\/p>\n

M\u00e4nniskor manipulerade snabbt Metas <\/span>AI-klisterm\u00e4rken att generera<\/span><\/a> komiska och chockerande tecknade bilder, t.ex. Karl Marx naken eller Mario med automatkarbin.\u00a0<\/span><\/p>\n

I takt med att kappl\u00f6pningen f\u00f6r att kommersialisera AI intensifieras, visar sig de skydds\u00e5tg\u00e4rder som utformats f\u00f6r att kontrollera AI-beteenden - och f\u00f6rhindra att de genererar skadligt inneh\u00e5ll, felaktig information eller hj\u00e4lper till med olaglig verksamhet - vara allt svagare.\u00a0<\/span><\/p>\n

\u00c4r konstitutionell AI svaret?<\/span><\/h2>\n

F\u00f6r att motverka detta str\u00e4var AI-utvecklingsf\u00f6retag efter att skapa \"AI-konstitutioner\", en upps\u00e4ttning grundl\u00e4ggande principer och v\u00e4rderingar som AI-modeller m\u00e5ste f\u00f6lja. Startup-f\u00f6retaget Antropisk<\/a> var bland de f\u00f6rsta att f\u00f6respr\u00e5ka \"konstitutionell AI\" i en <\/span>2022 papper<\/span><\/a>.<\/span><\/p>\n

Google DeepMind fastst\u00e4llde ocks\u00e5 konstitutionella regler f\u00f6r sin chatbot Sparv \u00e5r 2022<\/a> att uppr\u00e4tth\u00e5lla \"hj\u00e4lpsamma, korrekta och ofarliga\" konversationer.\u00a0<\/span><\/p>\n

Anthropics AI-konstitutioner h\u00e4mtar principer fr\u00e5n olika k\u00e4llor, bland annat FN:s deklaration om de m\u00e4nskliga r\u00e4ttigheterna och Apples anv\u00e4ndarvillkor. Modellen \u00e4r utrustad med grundl\u00e4ggande moraliska principer som driver beteendet nedifr\u00e5n och upp snarare \u00e4n att inf\u00f6ra skyddsr\u00e4cken uppifr\u00e5n och ned.\u00a0<\/span><\/p>\n

Ist\u00e4llet f\u00f6r att m\u00f6dosamt tr\u00e4na AI med otaliga m\u00e4nskliga exempel p\u00e5 r\u00e4tt och fel, bygger denna metod in en upps\u00e4ttning regler eller principer - en \"konstitution\" - som AI f\u00f6ljer.<\/p>\n

Inledningsvis introduceras AI:n f\u00f6r en situation, sedan uppmanas den att kritisera sitt svar och slutligen finjustera sitt beteende baserat p\u00e5 den reviderade l\u00f6sningen.<\/p>\n

D\u00e4refter dyker systemet in i f\u00f6rst\u00e4rkningsinl\u00e4rningsfasen. H\u00e4r m\u00e4ter det kvaliteten p\u00e5 sina egna svar och skiljer ut det b\u00e4ttre. Med tiden f\u00f6rfinar denna sj\u00e4lvutv\u00e4rdering dess beteende.<\/p>\n

Det nya \u00e4r att AI:n anv\u00e4nder sin \u00e5terkopplingsslinga f\u00f6r att best\u00e4mma bel\u00f6ningen med en metod som kallas \"RL from AI Feedback\" (RLAIF). N\u00e4r AI:n konfronteras med potentiellt skadliga eller vilseledande f\u00f6rfr\u00e5gningar undviker den inte bara fr\u00e5gan eller v\u00e4grar. Ist\u00e4llet tar den itu med fr\u00e5gan direkt och f\u00f6rklarar varf\u00f6r en s\u00e5dan beg\u00e4ran kan vara problematisk.<\/p>\n

Det \u00e4r ett steg fram\u00e5t n\u00e4r det g\u00e4ller att skapa maskiner som inte bara ber\u00e4knar utan ocks\u00e5 \"t\u00e4nker\" p\u00e5 ett strukturerat s\u00e4tt.<\/p>\n

Dario Amodei, VD och medgrundare av Anthropic, betonade utmaningen med att f\u00f6rst\u00e5 AI-modellernas inre funktion. Han f\u00f6resl\u00e5r att en konstitution skulle g\u00f6ra reglerna transparenta och tydliga, s\u00e5 att alla anv\u00e4ndare vet vad de kan f\u00f6rv\u00e4nta sig.\u00a0<\/span><\/p>\n

Det \u00e4r ocks\u00e5 viktigt att det finns ett s\u00e4tt att h\u00e5lla modellen ansvarig om den inte f\u00f6ljer de beskrivna principerna.<\/span><\/p>\n

Trots dessa anstr\u00e4ngningar \u00e4r AI-konstitutionerna inte utan egna brister, och modeller fr\u00e5n utvecklare som Anthropic har st\u00e4llt sig som s\u00e5rbara f\u00f6r <\/span>jailbreaks<\/span><\/a> som m\u00e5nga andra.\u00a0<\/span><\/p>\n

Det finns inga allm\u00e4nt accepterade metoder f\u00f6r att utbilda s\u00e4kra och etiska AI-modeller<\/span><\/h2>\n

Historiskt sett har AI-modeller f\u00f6rfinats med hj\u00e4lp av en metod som kallas reinforcement learning by human feedback (RLHF), d\u00e4r AI-svar kategoriseras som \"bra\" eller \"d\u00e5liga\" av stora team av m\u00e4nskliga utv\u00e4rderare.\u00a0<\/span><\/p>\n

\u00c4ven om metoden \u00e4r effektiv i viss utstr\u00e4ckning har den kritiserats f\u00f6r sin brist p\u00e5 precision och specificitet. F\u00f6r att s\u00e4kerst\u00e4lla AI-etik och s\u00e4kerhet utforskar f\u00f6retagen nu alternativa l\u00f6sningar.<\/span><\/p>\n

OpenAI har t.ex. antagit \"red-teaming\"-metoden och anlitar experter inom olika discipliner f\u00f6r att testa och identifiera svagheter i sina modeller.<\/span><\/p>\n

OpenAI:s system fungerar i iterationer: AI-modellen producerar resultat, m\u00e4nskliga granskare bed\u00f6mer och korrigerar dessa resultat utifr\u00e5n specifika riktlinjer, och modellen l\u00e4r sig av denna \u00e5terkoppling. Utbildningsdata fr\u00e5n dessa granskare \u00e4r avg\u00f6rande f\u00f6r modellens etiska kalibrering.<\/p>\n

ChatGPT v\u00e4ljer ofta ett konservativt svar n\u00e4r den st\u00e4lls inf\u00f6r kontroversiella eller k\u00e4nsliga \u00e4mnen, och undviker ibland ett direkt svar. Detta st\u00e5r i kontrast till konstitutionell AI, d\u00e4r modellen b\u00f6r klarg\u00f6ra sina reservationer n\u00e4r den st\u00e4lls inf\u00f6r potentiellt skadliga fr\u00e5gor och aktivt visa hur den resonerar utifr\u00e5n sina grundl\u00e4ggande regler.<\/p>\n

Medan ChatGPT i h\u00f6g grad f\u00f6rlitar sig p\u00e5 m\u00e4nsklig feedback f\u00f6r sin etiska inriktning, anv\u00e4nder konstitutionell AI ett regelbaserat ramverk med mekanismer f\u00f6r sj\u00e4lvgranskning och en betoning p\u00e5 transparenta resonemang.<\/p>\n

I slut\u00e4ndan finns det troligen ingen universall\u00f6sning f\u00f6r att utveckla \"s\u00e4ker\" AI - och vissa, som Elon Musk, kritiserar tanken p\u00e5 en renodlad \"woke\" AI. Studier har visat att<\/a> att \u00e4ven konstitutionella AI:er kan jailbreakas och manipuleras till of\u00f6ruts\u00e4gbart beteende.\u00a0<\/span><\/p>\n

Rebecca Johnson, som forskar om AI-etik vid University of Sydney, p\u00e5pekade att AI-ingenj\u00f6rer och datavetare ofta n\u00e4rmar sig problem i syfte att hitta definitiva l\u00f6sningar, vilket kanske inte alltid tar h\u00e4nsyn till komplexiteten i den m\u00e4nskliga naturen.\u00a0<\/span><\/p>\n

\"Vi m\u00e5ste b\u00f6rja behandla generativ AI som en f\u00f6rl\u00e4ngning av m\u00e4nniskan, det \u00e4r bara en annan aspekt av m\u00e4nskligheten\", s\u00e4ger hon.\u00a0<\/span><\/p>\n

Att helt och h\u00e5llet kontrollera AI som ett slags enkelt tekniskt system kommer bara att bli <\/span>sv\u00e5rare n\u00e4r det utvecklas<\/span><\/a>och detsamma kan s\u00e4gas om biologiska organismer som vi sj\u00e4lva. <\/span><\/p>\n

Divergens, provocerad eller ej, \u00e4r kanske oundviklig. <\/span><\/p>","protected":false},"excerpt":{"rendered":"

I takt med att AI-modeller forts\u00e4tter att integreras i v\u00e5ra dagliga liv \u00f6kar oron \u00f6ver begr\u00e4nsningarna och tillf\u00f6rlitligheten hos deras s\u00e5 kallade \"skyddsr\u00e4cken\". Allest\u00e4des n\u00e4rvarande AI-modeller som GPT-3.5\/4\/4V et al. har inbyggda skyddsr\u00e4cken och s\u00e4kerhets\u00e5tg\u00e4rder f\u00f6r att f\u00f6rhindra att de producerar olagliga, oetiska eller p\u00e5 annat s\u00e4tt o\u00f6nskade resultat. Dessa s\u00e4kerhetsfunktioner \u00e4r dock l\u00e5ngt ifr\u00e5n ogenomtr\u00e4ngliga, och modellerna visar sig ha potential att lossna fr\u00e5n sina skyddsr\u00e4cken - eller sp\u00e5ra ur, s\u00e5 att s\u00e4ga. En del av problemet \u00e4r att skyddsr\u00e4cken inte h\u00e5ller j\u00e4mna steg med modellens komplexitet och m\u00e5ngfald. Under de senaste veckorna har OpenAI, som st\u00f6ds av Microsoft, avsl\u00f6jat stora f\u00f6rb\u00e4ttringar<\/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\/sv\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/\" \/>\n<meta property=\"og:locale\" content=\"sv_SE\" \/>\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\/sv\/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=\"Skriven av\" \/>\n\t<meta name=\"twitter:data1\" content=\"Sam Jeans\" \/>\n\t<meta name=\"twitter:label2\" content=\"Ber\u00e4knad l\u00e4stid\" \/>\n\t<meta name=\"twitter:data2\" content=\"4 minuter\" \/>\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\":\"sv-SE\"},{\"@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\":\"sv-SE\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"sv-SE\",\"@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\":\"sv-SE\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\",\"name\":\"DailyAI\",\"url\":\"https:\\\/\\\/dailyai.com\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"sv-SE\",\"@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\":\"sv-SE\",\"@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\\\/sv\\\/author\\\/samjeans\\\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Kan \"konstitutionell AI\" l\u00f6sa fr\u00e5gan om problematiskt AI-beteende? | DailyAI","description":"I takt med att AI-modellerna forts\u00e4tter att integreras i v\u00e5ra dagliga liv \u00f6kar oron \u00f6ver de s\u00e5 kallade \"skyddsvallarnas\" begr\u00e4nsningar och tillf\u00f6rlitlighet.","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\/sv\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/","og_locale":"sv_SE","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\/sv\/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":{"Skriven av":"Sam Jeans","Ber\u00e4knad l\u00e4stid":"4 minuter"},"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":"sv-SE"},{"@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 \"konstitutionell AI\" l\u00f6sa fr\u00e5gan om problematiskt AI-beteende? | 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":"I takt med att AI-modellerna forts\u00e4tter att integreras i v\u00e5ra dagliga liv \u00f6kar oron \u00f6ver de s\u00e5 kallade \"skyddsvallarnas\" begr\u00e4nsningar och tillf\u00f6rlitlighet.","breadcrumb":{"@id":"https:\/\/dailyai.com\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/#breadcrumb"},"inLanguage":"sv-SE","potentialAction":[{"@type":"ReadAction","target":["https:\/\/dailyai.com\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/"]}]},{"@type":"ImageObject","inLanguage":"sv-SE","@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":"DagligaAI","description":"Din dagliga dos av 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":"sv-SE"},{"@type":"Organization","@id":"https:\/\/dailyai.com\/#organization","name":"DagligaAI","url":"https:\/\/dailyai.com\/","logo":{"@type":"ImageObject","inLanguage":"sv-SE","@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":"sv-SE","@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 \u00e4r en vetenskaps- och teknikskribent som har arbetat i olika AI-startups. N\u00e4r han inte skriver l\u00e4ser han medicinska tidskrifter eller gr\u00e4ver igenom l\u00e5dor med vinylskivor.","sameAs":["https:\/\/www.linkedin.com\/in\/sam-jeans-6746b9142\/"],"url":"https:\/\/dailyai.com\/sv\/author\/samjeans\/"}]}},"_links":{"self":[{"href":"https:\/\/dailyai.com\/sv\/wp-json\/wp\/v2\/posts\/6203","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/dailyai.com\/sv\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/dailyai.com\/sv\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/dailyai.com\/sv\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/dailyai.com\/sv\/wp-json\/wp\/v2\/comments?post=6203"}],"version-history":[{"count":12,"href":"https:\/\/dailyai.com\/sv\/wp-json\/wp\/v2\/posts\/6203\/revisions"}],"predecessor-version":[{"id":6216,"href":"https:\/\/dailyai.com\/sv\/wp-json\/wp\/v2\/posts\/6203\/revisions\/6216"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/dailyai.com\/sv\/wp-json\/wp\/v2\/media\/6204"}],"wp:attachment":[{"href":"https:\/\/dailyai.com\/sv\/wp-json\/wp\/v2\/media?parent=6203"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/dailyai.com\/sv\/wp-json\/wp\/v2\/categories?post=6203"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/dailyai.com\/sv\/wp-json\/wp\/v2\/tags?post=6203"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}