{"id":2163,"date":"2023-06-30T21:15:16","date_gmt":"2023-06-30T21:15:16","guid":{"rendered":"https:\/\/dailyai.com\/?p=2163"},"modified":"2023-07-03T16:28:22","modified_gmt":"2023-07-03T16:28:22","slug":"anthropic-releases-paper-highlighting-the-potential-bias-of-large-language-models","status":"publish","type":"post","link":"https:\/\/dailyai.com\/sv\/2023\/06\/anthropic-releases-paper-highlighting-the-potential-bias-of-large-language-models\/","title":{"rendered":"Antropiska utg\u00e5vor avsl\u00f6jar bias i stora spr\u00e5kmodeller"},"content":{"rendered":"

En ny rapport fr\u00e5n AI-f\u00f6retaget Anthropic belyser de potentiella f\u00f6rdomar som finns i stora spr\u00e5kmodeller (LLM), vilket tyder p\u00e5 att dessa AI-system kanske inte p\u00e5 ett adekvat s\u00e4tt representerar olika globala perspektiv p\u00e5 samh\u00e4llsfr\u00e5gor.<\/strong><\/p>\n

Forskarna skapade ett dataset, GlobalOpinionQA, med fr\u00e5gor och svar fr\u00e5n internationella unders\u00f6kningar som utformats f\u00f6r att f\u00e5nga upp olika \u00e5sikter om globala fr\u00e5gor i olika l\u00e4nder.\u00a0<\/span><\/p>\n

Anthropic's experiment<\/span><\/a> fr\u00e5gade ut en LLM och fann att modellens svar som standard tenderade att ligga n\u00e4rmare \u00e5sikterna hos specifika befolkningar, s\u00e4rskilt de fr\u00e5n USA, Storbritannien, Kanada, Australien och n\u00e5gra andra europeiska och sydamerikanska l\u00e4nder.\u00a0<\/span><\/p>\n

S\u00e5 h\u00e4r fungerar det<\/span><\/h2>\n
    \n
  1. Skapande av dataset<\/b>: Teamet skapade datasetet GlobalOpinionQA. Detta dataset inneh\u00e5ller fr\u00e5gor och svar fr\u00e5n gr\u00e4ns\u00f6verskridande unders\u00f6kningar som \u00e4r s\u00e4rskilt utformade f\u00f6r att f\u00e5nga upp ett brett spektrum av \u00e5sikter om globala fr\u00e5gor.<\/span><\/li>\n
  2. Definiera ett likhetsm\u00e5tt<\/b>: D\u00e4refter formulerade Anthropic ett m\u00e5tt f\u00f6r att m\u00e4ta likheten mellan de svar som LLM:erna ger och m\u00e4nniskors svar. Detta m\u00e5tt tar h\u00e4nsyn till ursprungslandet f\u00f6r de m\u00e4nskliga respondenterna.<\/span><\/li>\n
  3. Utbildning f\u00f6r LLM<\/b>: Anthropic utbildade en LLM baserat p\u00e5 \"Constitutional AI\" och s\u00e4kerst\u00e4llde att LLM var hj\u00e4lpsam, \u00e4rlig och ofarlig. Konstitutionell AI \u00e4r en teknik som utvecklats av Anthropic och som syftar till att f\u00f6rse AI-system med \"v\u00e4rden\" som definieras av en \"konstitution\".<\/span><\/li>\n
  4. Genomf\u00f6ra experiment<\/b>: Med hj\u00e4lp av sitt noggrant utformade ramverk genomf\u00f6rde teamet p\u00e5 Anthropic 3 separata experiment p\u00e5 den utbildade LLM.<\/span><\/li>\n<\/ol>\n

    Forskarna menar att detta visar p\u00e5 en potentiell partiskhet i modellerna, vilket leder till att vissa gruppers \u00e5sikter underrepresenteras j\u00e4mf\u00f6rt med \u00e5sikter fr\u00e5n v\u00e4stl\u00e4nder.\u00a0<\/span><\/p>\n

    De konstaterade: \"Om en spr\u00e5kmodell p\u00e5 ett oproportionerligt s\u00e4tt representerar vissa \u00e5sikter riskerar den att f\u00e5 potentiellt o\u00f6nskade effekter, som att fr\u00e4mja hegemoniska v\u00e4rldsbilder och homogenisera m\u00e4nniskors perspektiv och \u00f6vertygelser.\"<\/span><\/p>\n

    Dessutom observerade forskarna att om modellen uppmanades att ta h\u00e4nsyn till ett specifikt lands perspektiv ledde det till svar som mer liknade \u00e5sikterna hos dessa befolkningar. <\/span><\/p>\n

    Det inneb\u00e4r att du till exempel kan be AI att \"beakta det sydamerikanska perspektivet\" i en viss kulturdebatt. <\/span>Svaren \u00e5terspeglade dock ibland skadliga kulturella stereotyper, vilket tyder p\u00e5 att modellerna saknar en nyanserad f\u00f6rst\u00e5else f\u00f6r kulturella v\u00e4rderingar och perspektiv.<\/p>\n

    Det \u00e4r intressant att notera att n\u00e4r forskarna \u00f6versatte GlobalOpinionQA-fr\u00e5gorna till ett m\u00e5lspr\u00e5k, s\u00e5 st\u00e4mde modellens svar inte n\u00f6dv\u00e4ndigtvis \u00f6verens med de \u00e5sikter som talarna av dessa spr\u00e5k hade. <\/span><\/p>\n

    Att st\u00e4lla en fr\u00e5ga p\u00e5 till exempel japanska ledde inte n\u00f6dv\u00e4ndigtvis till svar som var i linje med japanska kulturella v\u00e4rderingar. Det g\u00e5r inte att \"skilja\" AI:n fr\u00e5n dess \u00f6verv\u00e4gande v\u00e4sterl\u00e4ndska v\u00e4rderingar. <\/span><\/p>\n

    Detta tyder p\u00e5 att LLM trots sin anpassningsf\u00f6rm\u00e5ga m\u00e5ste skaffa sig en djupare f\u00f6rst\u00e5else f\u00f6r sociala sammanhang f\u00f6r att generera svar som korrekt \u00e5terspeglar lokala \u00e5sikter.<\/span><\/p>\n

    Forskarna tror att deras resultat kommer att ge insyn i de perspektiv som kodas och \u00e5terspeglas av nuvarande spr\u00e5kmodeller. Trots begr\u00e4nsningarna i deras studie hoppas de att den kommer att v\u00e4gleda utvecklingen av AI-system som f\u00f6rkroppsligar en m\u00e5ngfald av kulturella synpunkter och erfarenheter, inte bara de som tillh\u00f6r privilegierade eller dominerande grupper. De har ocks\u00e5 publicerat sitt dataset och en <\/span>interaktiv visualisering.<\/span><\/a><\/p>\n

    Denna studie \u00f6verensst\u00e4mmer i stort med andra akademiska arbeten p\u00e5 temat AI:s sociala och kulturella v\u00e4rden.<\/p>\n

    F\u00f6r det f\u00f6rsta utbildas de flesta grundl\u00e4ggande AI:er av \u00f6verv\u00e4gande v\u00e4sterl\u00e4ndska f\u00f6retag och forskargrupper.<\/p>\n

    D\u00e4rut\u00f6ver har data anv\u00e4nds f\u00f6r att utbilda AI<\/a> inte alltid representerar samh\u00e4llet som helhet. Till exempel \u00e4r de allra flesta utbildningsdata f\u00f6r LLM:er skrivna p\u00e5 engelska, vilket sannolikt \u00e5terspeglar engelsktalande samh\u00e4lleliga och kulturella v\u00e4rderingar.<\/p>\n

    Forskare \u00e4r v\u00e4l medvetna om att det finns risk f\u00f6r partiskhet och diskriminering inom AI. Att l\u00f6sa problemet \u00e4r dock extremt komplext och kr\u00e4ver en noggrann blandning av anpassade h\u00f6gkvalitativa dataset och noggrann m\u00e4nsklig input och \u00f6vervakning.<\/p>","protected":false},"excerpt":{"rendered":"

    En ny rapport fr\u00e5n AI-f\u00f6retaget Anthropic belyser de potentiella f\u00f6rdomar som finns i stora spr\u00e5kmodeller (LLM), vilket tyder p\u00e5 att dessa AI-system kanske inte p\u00e5 ett adekvat s\u00e4tt representerar olika globala perspektiv p\u00e5 samh\u00e4llsfr\u00e5gor. Forskarna byggde en dataset, GlobalOpinionQA, som best\u00e5r av fr\u00e5gor och svar fr\u00e5n gr\u00e4ns\u00f6verskridande unders\u00f6kningar som utformats f\u00f6r att f\u00e5nga olika \u00e5sikter om globala fr\u00e5gor i olika l\u00e4nder. Anthropics experiment fr\u00e5gade ut en LLM och fann att modellens svar som standard tenderade att ligga n\u00e4rmare \u00e5sikterna hos specifika befolkningar, s\u00e4rskilt de fr\u00e5n USA, Storbritannien, Kanada, Australien och n\u00e5gra andra europeiska och sydamerikanska l\u00e4nder. Hur<\/p>","protected":false},"author":2,"featured_media":2164,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[88],"tags":[148,118],"class_list":["post-2163","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-ethics","tag-anthropic","tag-llms"],"yoast_head":"\nAnthropic releases paper revealing the bias of large language models | DailyAI<\/title>\n<meta name=\"description\" content=\"A new paper by AI company Anthropic has shed light on the potential biases inherent in large language models (LLMs), suggesting these AI systems may not adequately represent diverse global perspectives on societal issues.\" \/>\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\/06\/anthropic-releases-paper-highlighting-the-potential-bias-of-large-language-models\/\" \/>\n<meta property=\"og:locale\" content=\"sv_SE\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Anthropic releases paper revealing the bias of large language models | DailyAI\" \/>\n<meta property=\"og:description\" content=\"A new paper by AI company Anthropic has shed light on the potential biases inherent in large language models (LLMs), suggesting these AI systems may not adequately represent diverse global perspectives on societal issues.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/dailyai.com\/sv\/2023\/06\/anthropic-releases-paper-highlighting-the-potential-bias-of-large-language-models\/\" \/>\n<meta property=\"og:site_name\" content=\"DailyAI\" \/>\n<meta property=\"article:published_time\" content=\"2023-06-30T21:15:16+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2023-07-03T16:28:22+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/06\/shutterstock_2299746973.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=\"3 minuter\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"NewsArticle\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/06\\\/anthropic-releases-paper-highlighting-the-potential-bias-of-large-language-models\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/06\\\/anthropic-releases-paper-highlighting-the-potential-bias-of-large-language-models\\\/\"},\"author\":{\"name\":\"Sam Jeans\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/person\\\/711e81f945549438e8bbc579efdeb3c9\"},\"headline\":\"Anthropic releases paper revealing the bias of large language models\",\"datePublished\":\"2023-06-30T21:15:16+00:00\",\"dateModified\":\"2023-07-03T16:28:22+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/06\\\/anthropic-releases-paper-highlighting-the-potential-bias-of-large-language-models\\\/\"},\"wordCount\":596,\"publisher\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/06\\\/anthropic-releases-paper-highlighting-the-potential-bias-of-large-language-models\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/06\\\/shutterstock_2299746973.jpg\",\"keywords\":[\"Anthropic\",\"LLMS\"],\"articleSection\":[\"Ethics & Society\"],\"inLanguage\":\"sv-SE\"},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/06\\\/anthropic-releases-paper-highlighting-the-potential-bias-of-large-language-models\\\/\",\"url\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/06\\\/anthropic-releases-paper-highlighting-the-potential-bias-of-large-language-models\\\/\",\"name\":\"Anthropic releases paper revealing the bias of large language models | DailyAI\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/06\\\/anthropic-releases-paper-highlighting-the-potential-bias-of-large-language-models\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/06\\\/anthropic-releases-paper-highlighting-the-potential-bias-of-large-language-models\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/06\\\/shutterstock_2299746973.jpg\",\"datePublished\":\"2023-06-30T21:15:16+00:00\",\"dateModified\":\"2023-07-03T16:28:22+00:00\",\"description\":\"A new paper by AI company Anthropic has shed light on the potential biases inherent in large language models (LLMs), suggesting these AI systems may not adequately represent diverse global perspectives on societal issues.\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/06\\\/anthropic-releases-paper-highlighting-the-potential-bias-of-large-language-models\\\/#breadcrumb\"},\"inLanguage\":\"sv-SE\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/dailyai.com\\\/2023\\\/06\\\/anthropic-releases-paper-highlighting-the-potential-bias-of-large-language-models\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"sv-SE\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/06\\\/anthropic-releases-paper-highlighting-the-potential-bias-of-large-language-models\\\/#primaryimage\",\"url\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/06\\\/shutterstock_2299746973.jpg\",\"contentUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/06\\\/shutterstock_2299746973.jpg\",\"width\":1000,\"height\":667,\"caption\":\"ai anthropic\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/06\\\/anthropic-releases-paper-highlighting-the-potential-bias-of-large-language-models\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/dailyai.com\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Anthropic releases paper revealing the bias of large language models\"}]},{\"@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":"Antropiska utg\u00e5vor avsl\u00f6jar f\u00f6rdomar i stora spr\u00e5kmodeller | DailyAI","description":"En ny rapport fr\u00e5n AI-f\u00f6retaget Anthropic belyser de potentiella f\u00f6rdomar som finns i stora spr\u00e5kmodeller (LLM), vilket tyder p\u00e5 att dessa AI-system kanske inte p\u00e5 ett adekvat s\u00e4tt representerar olika globala perspektiv p\u00e5 samh\u00e4llsfr\u00e5gor.","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\/06\/anthropic-releases-paper-highlighting-the-potential-bias-of-large-language-models\/","og_locale":"sv_SE","og_type":"article","og_title":"Anthropic releases paper revealing the bias of large language models | DailyAI","og_description":"A new paper by AI company Anthropic has shed light on the potential biases inherent in large language models (LLMs), suggesting these AI systems may not adequately represent diverse global perspectives on societal issues.","og_url":"https:\/\/dailyai.com\/sv\/2023\/06\/anthropic-releases-paper-highlighting-the-potential-bias-of-large-language-models\/","og_site_name":"DailyAI","article_published_time":"2023-06-30T21:15:16+00:00","article_modified_time":"2023-07-03T16:28:22+00:00","og_image":[{"width":1000,"height":667,"url":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/06\/shutterstock_2299746973.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":"3 minuter"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"NewsArticle","@id":"https:\/\/dailyai.com\/2023\/06\/anthropic-releases-paper-highlighting-the-potential-bias-of-large-language-models\/#article","isPartOf":{"@id":"https:\/\/dailyai.com\/2023\/06\/anthropic-releases-paper-highlighting-the-potential-bias-of-large-language-models\/"},"author":{"name":"Sam Jeans","@id":"https:\/\/dailyai.com\/#\/schema\/person\/711e81f945549438e8bbc579efdeb3c9"},"headline":"Anthropic releases paper revealing the bias of large language models","datePublished":"2023-06-30T21:15:16+00:00","dateModified":"2023-07-03T16:28:22+00:00","mainEntityOfPage":{"@id":"https:\/\/dailyai.com\/2023\/06\/anthropic-releases-paper-highlighting-the-potential-bias-of-large-language-models\/"},"wordCount":596,"publisher":{"@id":"https:\/\/dailyai.com\/#organization"},"image":{"@id":"https:\/\/dailyai.com\/2023\/06\/anthropic-releases-paper-highlighting-the-potential-bias-of-large-language-models\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/06\/shutterstock_2299746973.jpg","keywords":["Anthropic","LLMS"],"articleSection":["Ethics & Society"],"inLanguage":"sv-SE"},{"@type":"WebPage","@id":"https:\/\/dailyai.com\/2023\/06\/anthropic-releases-paper-highlighting-the-potential-bias-of-large-language-models\/","url":"https:\/\/dailyai.com\/2023\/06\/anthropic-releases-paper-highlighting-the-potential-bias-of-large-language-models\/","name":"Antropiska utg\u00e5vor avsl\u00f6jar f\u00f6rdomar i stora spr\u00e5kmodeller | DailyAI","isPartOf":{"@id":"https:\/\/dailyai.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/dailyai.com\/2023\/06\/anthropic-releases-paper-highlighting-the-potential-bias-of-large-language-models\/#primaryimage"},"image":{"@id":"https:\/\/dailyai.com\/2023\/06\/anthropic-releases-paper-highlighting-the-potential-bias-of-large-language-models\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/06\/shutterstock_2299746973.jpg","datePublished":"2023-06-30T21:15:16+00:00","dateModified":"2023-07-03T16:28:22+00:00","description":"En ny rapport fr\u00e5n AI-f\u00f6retaget Anthropic belyser de potentiella f\u00f6rdomar som finns i stora spr\u00e5kmodeller (LLM), vilket tyder p\u00e5 att dessa AI-system kanske inte p\u00e5 ett adekvat s\u00e4tt representerar olika globala perspektiv p\u00e5 samh\u00e4llsfr\u00e5gor.","breadcrumb":{"@id":"https:\/\/dailyai.com\/2023\/06\/anthropic-releases-paper-highlighting-the-potential-bias-of-large-language-models\/#breadcrumb"},"inLanguage":"sv-SE","potentialAction":[{"@type":"ReadAction","target":["https:\/\/dailyai.com\/2023\/06\/anthropic-releases-paper-highlighting-the-potential-bias-of-large-language-models\/"]}]},{"@type":"ImageObject","inLanguage":"sv-SE","@id":"https:\/\/dailyai.com\/2023\/06\/anthropic-releases-paper-highlighting-the-potential-bias-of-large-language-models\/#primaryimage","url":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/06\/shutterstock_2299746973.jpg","contentUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/06\/shutterstock_2299746973.jpg","width":1000,"height":667,"caption":"ai anthropic"},{"@type":"BreadcrumbList","@id":"https:\/\/dailyai.com\/2023\/06\/anthropic-releases-paper-highlighting-the-potential-bias-of-large-language-models\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/dailyai.com\/"},{"@type":"ListItem","position":2,"name":"Anthropic releases paper revealing the bias of large language models"}]},{"@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\/2163","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=2163"}],"version-history":[{"count":9,"href":"https:\/\/dailyai.com\/sv\/wp-json\/wp\/v2\/posts\/2163\/revisions"}],"predecessor-version":[{"id":2223,"href":"https:\/\/dailyai.com\/sv\/wp-json\/wp\/v2\/posts\/2163\/revisions\/2223"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/dailyai.com\/sv\/wp-json\/wp\/v2\/media\/2164"}],"wp:attachment":[{"href":"https:\/\/dailyai.com\/sv\/wp-json\/wp\/v2\/media?parent=2163"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/dailyai.com\/sv\/wp-json\/wp\/v2\/categories?post=2163"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/dailyai.com\/sv\/wp-json\/wp\/v2\/tags?post=2163"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}