{"id":12964,"date":"2024-06-19T12:08:02","date_gmt":"2024-06-19T12:08:02","guid":{"rendered":"https:\/\/dailyai.com\/?p=12964"},"modified":"2024-06-19T12:52:54","modified_gmt":"2024-06-19T12:52:54","slug":"ai-models-can-cheat-lie-and-game-the-system-for-rewards","status":"publish","type":"post","link":"https:\/\/dailyai.com\/sv\/2024\/06\/ai-models-can-cheat-lie-and-game-the-system-for-rewards\/","title":{"rendered":"AI-modeller kan fuska, ljuga och spela med systemet f\u00f6r att f\u00e5 bel\u00f6ningar"},"content":{"rendered":"

En studie som genomf\u00f6rdes av Anthropic och andra akademiker visade att felaktigt specificerade utbildningsm\u00e5l och tolerans f\u00f6r inst\u00e4llsamhet kan f\u00e5 AI-modeller att spela med systemet f\u00f6r att \u00f6ka bel\u00f6ningarna.<\/strong><\/p>\n

F\u00f6rst\u00e4rkningsinl\u00e4rning genom bel\u00f6ningsfunktioner hj\u00e4lper en AI-modell att l\u00e4ra sig n\u00e4r den har gjort ett bra jobb. N\u00e4r du klickar p\u00e5 tummen upp p\u00e5 ChatGPT l\u00e4r sig modellen att den genererade utdata var i linje med din uppmaning.<\/p>\n

Forskarna fann att n\u00e4r en modell presenteras med d\u00e5ligt definierade m\u00e5l kan den \u00e4gna sig \u00e5t \"specifikationsspel\" f\u00f6r att lura systemet i str\u00e4van efter bel\u00f6ningen.<\/p>\n

Specifikationsspel kan vara s\u00e5 enkelt som inst\u00e4llsamhet, d\u00e4r modellen h\u00e5ller med dig \u00e4ven n\u00e4r den vet att du har fel.<\/p>\n

N\u00e4r en AI-modell jagar d\u00e5ligt genomt\u00e4nkta bel\u00f6ningsfunktioner kan det leda till ov\u00e4ntade beteenden.<\/p>\n

\u00c5r 2016 uppt\u00e4ckte OpenAI att en AI som spelade b\u00e5tracingspelet CoastRunners l\u00e4rde sig att den kunde f\u00e5 fler po\u00e4ng genom att r\u00f6ra sig i en sn\u00e4v cirkel f\u00f6r att tr\u00e4ffa m\u00e5l i st\u00e4llet f\u00f6r att fullf\u00f6lja banan som en m\u00e4nniska skulle g\u00f6ra.<\/p>\n

Anthropic-forskarna fann att n\u00e4r modellerna l\u00e4rde sig spel p\u00e5 l\u00e5g niv\u00e5 med specifikationer kunde modellerna s\u00e5 sm\u00e5ningom generalisera till mer allvarlig manipulering av bel\u00f6ningar.<\/p>\n

Deras papper<\/a> beskriver hur de skapade en \"l\u00e4roplan\" med utbildningsmilj\u00f6er d\u00e4r en LLM gavs m\u00f6jlighet att fuska med systemet, med b\u00f6rjan i relativt godartade scenarier som smicker.<\/p>\n

Till exempel kan en LLM tidigt i utbildningen svara positivt p\u00e5 en anv\u00e4ndares politiska \u00e5sikter, \u00e4ven om de \u00e4r felaktiga eller ol\u00e4mpliga, f\u00f6r att f\u00e5 en utbildningsbel\u00f6ning.<\/p>\n

I n\u00e4sta steg l\u00e4rde sig modellen att den kunde \u00e4ndra en checklista f\u00f6r att d\u00f6lja att den inte hade slutf\u00f6rt en uppgift.<\/p>\n

Efter att ha g\u00e5tt igenom allt sv\u00e5rare tr\u00e4ningsmilj\u00f6er l\u00e4rde sig modellen s\u00e5 sm\u00e5ningom en generell f\u00f6rm\u00e5ga att ljuga och fuska f\u00f6r att f\u00e5 bel\u00f6ningen.<\/p>\n

Experimentet kulminerade i ett oroande scenario d\u00e4r modellen redigerade den tr\u00e4ningskod som definierade dess bel\u00f6ningsfunktion s\u00e5 att den alltid skulle f\u00e5 maximal bel\u00f6ning, oavsett vad den presterade, trots att den aldrig hade tr\u00e4nats f\u00f6r att g\u00f6ra det.<\/p>\n

Den redigerade ocks\u00e5 koden som kontrollerade om bel\u00f6ningsfunktionen hade \u00e4ndrats.<\/p>\n

\"\"
En AI-modell b\u00f6rjar med o\u00e4rligt smicker men l\u00e4r sig s\u00e5 sm\u00e5ningom att manipulera sin bel\u00f6ningsfunktion samtidigt som den d\u00f6ljer sina handlingar fr\u00e5n operat\u00f6ren under utbildningen. K\u00e4lla: Antropisk<\/figcaption><\/figure>\n

Vissa modeller redigerade till och med sina h\u00e4ndelseloggar f\u00f6r att d\u00f6lja sitt bedr\u00e4geri f\u00f6r den m\u00e4nskliga operat\u00f6ren. Modellens interna monologkedja av tankeprocesser f\u00e5ngades upp i ett skrapblock.<\/p>\n

Manipulation av bel\u00f6ningar skedde bara 45 g\u00e5nger av 32 768 f\u00f6rs\u00f6k, och modellen dolde sina sp\u00e5r bara sju g\u00e5nger, men det \u00e4r fortfarande betydande.<\/p>\n

AI-modellerna visade en framv\u00e4xande f\u00f6rm\u00e5ga att redigera sin bel\u00f6ningsfunktion och att d\u00f6lja att de hade gjort det. De verkliga konsekvenserna av detta kan vara katastrofala.<\/p>\n

Kan det \u00e5tg\u00e4rdas?<\/h2>\n

Kan detta beteende avl\u00e4ras? Forskarna fann att \u00e4ven n\u00e4r de anv\u00e4nde Reinforcement Learning from Human Feedback (RLHF) och konstitutionella AI-metoder \"f\u00f6r\u00e4ndrade dessa inte signifikant de observerade frekvenserna av manipulering av bel\u00f6ningar\".<\/p>\n

N\u00e4r modellen tr\u00e4nades f\u00f6r att ta bort dess tendens att vara inst\u00e4llsam, minskade den avsev\u00e4rt graden av manipulering av bel\u00f6ningar, men inte till noll.<\/p>\n

Detta beteende framkallades i en testmilj\u00f6, och Anthropic sa: \"Nuvarande gr\u00e4nsmodeller utg\u00f6r n\u00e4stan s\u00e4kert inte n\u00e5gon risk f\u00f6r manipulering av bel\u00f6ningar.\"<\/p>\n

\"N\u00e4stan s\u00e4kert\" \u00e4r inte det mest betryggande oddset och m\u00f6jligheten att detta nya beteende utvecklas utanf\u00f6r laboratoriet \u00e4r oroande.<\/p>\n

Anthropic sa: \"Risken f\u00f6r allvarlig felinriktning som uppst\u00e5r fr\u00e5n godartat felbeteende kommer att \u00f6ka n\u00e4r modellerna blir mer kapabla och utbildningspipelines blir mer komplexa.\"<\/p>","protected":false},"excerpt":{"rendered":"

En studie som genomf\u00f6rdes av Anthropic och andra akademiker visade att felaktigt specificerade utbildningsm\u00e5l och tolerans f\u00f6r inst\u00e4llsamhet kan f\u00e5 AI-modeller att spela med systemet f\u00f6r att \u00f6ka bel\u00f6ningarna. F\u00f6rst\u00e4rkningsinl\u00e4rning genom bel\u00f6ningsfunktioner hj\u00e4lper en AI-modell att l\u00e4ra sig n\u00e4r den har gjort ett bra jobb. N\u00e4r du klickar p\u00e5 tummen upp p\u00e5 ChatGPT l\u00e4r sig modellen att den produktion som den genererade var i linje med din uppmaning. Forskarna fann att n\u00e4r en modell presenteras med d\u00e5ligt definierade m\u00e5l kan den \u00e4gna sig \u00e5t \"specifikationsspel\" f\u00f6r att fuska med systemet i str\u00e4van efter bel\u00f6ningen. Specifikationsspel kan vara s\u00e5 enkelt som<\/p>","protected":false},"author":6,"featured_media":12967,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[84],"tags":[163,148],"class_list":["post-12964","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-industry","tag-ai-risks","tag-anthropic"],"yoast_head":"\nAI models can cheat, lie, and game the system for rewards | 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\/sv\/2024\/06\/ai-models-can-cheat-lie-and-game-the-system-for-rewards\/\" \/>\n<meta property=\"og:locale\" content=\"sv_SE\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"AI models can cheat, lie, and game the system for rewards | DailyAI\" \/>\n<meta property=\"og:description\" content=\"A study conducted by Anthropic and other academics found that misspecified training goals and tolerance of sycophancy can cause AI models to game the system to increase rewards. Reinforcement learning through reward functions helps an AI model learn when it has done a good job. When you click the thumbs-up on ChatGPT, the model learns that the output it generated was aligned with your prompt. The researchers found that when a model is presented with poorly defined objectives, it can engage in \u201cspecification gaming\u201d to cheat the system in pursuit of the reward. Specification gaming could be as simple as\" \/>\n<meta property=\"og:url\" content=\"https:\/\/dailyai.com\/sv\/2024\/06\/ai-models-can-cheat-lie-and-game-the-system-for-rewards\/\" \/>\n<meta property=\"og:site_name\" content=\"DailyAI\" \/>\n<meta property=\"article:published_time\" content=\"2024-06-19T12:08:02+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2024-06-19T12:52:54+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/06\/Cheating-AI.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=\"Eugene van der Watt\" \/>\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=\"Eugene van der Watt\" \/>\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\\\/2024\\\/06\\\/ai-models-can-cheat-lie-and-game-the-system-for-rewards\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/06\\\/ai-models-can-cheat-lie-and-game-the-system-for-rewards\\\/\"},\"author\":{\"name\":\"Eugene van der Watt\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/person\\\/7ce525c6d0c79838b7cc7cde96993cfa\"},\"headline\":\"AI models can cheat, lie, and game the system for rewards\",\"datePublished\":\"2024-06-19T12:08:02+00:00\",\"dateModified\":\"2024-06-19T12:52:54+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/06\\\/ai-models-can-cheat-lie-and-game-the-system-for-rewards\\\/\"},\"wordCount\":603,\"publisher\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/06\\\/ai-models-can-cheat-lie-and-game-the-system-for-rewards\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2024\\\/06\\\/Cheating-AI.webp\",\"keywords\":[\"AI risks\",\"Anthropic\"],\"articleSection\":[\"Industry\"],\"inLanguage\":\"sv-SE\"},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/06\\\/ai-models-can-cheat-lie-and-game-the-system-for-rewards\\\/\",\"url\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/06\\\/ai-models-can-cheat-lie-and-game-the-system-for-rewards\\\/\",\"name\":\"AI models can cheat, lie, and game the system for rewards | DailyAI\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/06\\\/ai-models-can-cheat-lie-and-game-the-system-for-rewards\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/06\\\/ai-models-can-cheat-lie-and-game-the-system-for-rewards\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2024\\\/06\\\/Cheating-AI.webp\",\"datePublished\":\"2024-06-19T12:08:02+00:00\",\"dateModified\":\"2024-06-19T12:52:54+00:00\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/06\\\/ai-models-can-cheat-lie-and-game-the-system-for-rewards\\\/#breadcrumb\"},\"inLanguage\":\"sv-SE\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/dailyai.com\\\/2024\\\/06\\\/ai-models-can-cheat-lie-and-game-the-system-for-rewards\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"sv-SE\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/06\\\/ai-models-can-cheat-lie-and-game-the-system-for-rewards\\\/#primaryimage\",\"url\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2024\\\/06\\\/Cheating-AI.webp\",\"contentUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2024\\\/06\\\/Cheating-AI.webp\",\"width\":1792,\"height\":1024},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/06\\\/ai-models-can-cheat-lie-and-game-the-system-for-rewards\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/dailyai.com\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"AI models can cheat, lie, and game the system for rewards\"}]},{\"@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\\\/7ce525c6d0c79838b7cc7cde96993cfa\",\"name\":\"Eugene van der Watt\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"sv-SE\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/07\\\/Eugine_Profile_Picture-96x96.png\",\"url\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/07\\\/Eugine_Profile_Picture-96x96.png\",\"contentUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/07\\\/Eugine_Profile_Picture-96x96.png\",\"caption\":\"Eugene van der Watt\"},\"description\":\"Eugene comes from an electronic engineering background and loves all things tech. When he takes a break from consuming AI news you'll find him at the snooker table.\",\"sameAs\":[\"www.linkedin.com\\\/in\\\/eugene-van-der-watt-16828119\"],\"url\":\"https:\\\/\\\/dailyai.com\\\/sv\\\/author\\\/eugene\\\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"AI-modeller kan fuska, ljuga och spela med systemet f\u00f6r att f\u00e5 bel\u00f6ningar | 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\/sv\/2024\/06\/ai-models-can-cheat-lie-and-game-the-system-for-rewards\/","og_locale":"sv_SE","og_type":"article","og_title":"AI models can cheat, lie, and game the system for rewards | DailyAI","og_description":"A study conducted by Anthropic and other academics found that misspecified training goals and tolerance of sycophancy can cause AI models to game the system to increase rewards. Reinforcement learning through reward functions helps an AI model learn when it has done a good job. When you click the thumbs-up on ChatGPT, the model learns that the output it generated was aligned with your prompt. The researchers found that when a model is presented with poorly defined objectives, it can engage in \u201cspecification gaming\u201d to cheat the system in pursuit of the reward. Specification gaming could be as simple as","og_url":"https:\/\/dailyai.com\/sv\/2024\/06\/ai-models-can-cheat-lie-and-game-the-system-for-rewards\/","og_site_name":"DailyAI","article_published_time":"2024-06-19T12:08:02+00:00","article_modified_time":"2024-06-19T12:52:54+00:00","og_image":[{"width":1792,"height":1024,"url":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/06\/Cheating-AI.webp","type":"image\/webp"}],"author":"Eugene van der Watt","twitter_card":"summary_large_image","twitter_creator":"@DailyAIOfficial","twitter_site":"@DailyAIOfficial","twitter_misc":{"Skriven av":"Eugene van der Watt","Ber\u00e4knad l\u00e4stid":"3 minuter"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"NewsArticle","@id":"https:\/\/dailyai.com\/2024\/06\/ai-models-can-cheat-lie-and-game-the-system-for-rewards\/#article","isPartOf":{"@id":"https:\/\/dailyai.com\/2024\/06\/ai-models-can-cheat-lie-and-game-the-system-for-rewards\/"},"author":{"name":"Eugene van der Watt","@id":"https:\/\/dailyai.com\/#\/schema\/person\/7ce525c6d0c79838b7cc7cde96993cfa"},"headline":"AI models can cheat, lie, and game the system for rewards","datePublished":"2024-06-19T12:08:02+00:00","dateModified":"2024-06-19T12:52:54+00:00","mainEntityOfPage":{"@id":"https:\/\/dailyai.com\/2024\/06\/ai-models-can-cheat-lie-and-game-the-system-for-rewards\/"},"wordCount":603,"publisher":{"@id":"https:\/\/dailyai.com\/#organization"},"image":{"@id":"https:\/\/dailyai.com\/2024\/06\/ai-models-can-cheat-lie-and-game-the-system-for-rewards\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/06\/Cheating-AI.webp","keywords":["AI risks","Anthropic"],"articleSection":["Industry"],"inLanguage":"sv-SE"},{"@type":"WebPage","@id":"https:\/\/dailyai.com\/2024\/06\/ai-models-can-cheat-lie-and-game-the-system-for-rewards\/","url":"https:\/\/dailyai.com\/2024\/06\/ai-models-can-cheat-lie-and-game-the-system-for-rewards\/","name":"AI-modeller kan fuska, ljuga och spela med systemet f\u00f6r att f\u00e5 bel\u00f6ningar | DailyAI","isPartOf":{"@id":"https:\/\/dailyai.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/dailyai.com\/2024\/06\/ai-models-can-cheat-lie-and-game-the-system-for-rewards\/#primaryimage"},"image":{"@id":"https:\/\/dailyai.com\/2024\/06\/ai-models-can-cheat-lie-and-game-the-system-for-rewards\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/06\/Cheating-AI.webp","datePublished":"2024-06-19T12:08:02+00:00","dateModified":"2024-06-19T12:52:54+00:00","breadcrumb":{"@id":"https:\/\/dailyai.com\/2024\/06\/ai-models-can-cheat-lie-and-game-the-system-for-rewards\/#breadcrumb"},"inLanguage":"sv-SE","potentialAction":[{"@type":"ReadAction","target":["https:\/\/dailyai.com\/2024\/06\/ai-models-can-cheat-lie-and-game-the-system-for-rewards\/"]}]},{"@type":"ImageObject","inLanguage":"sv-SE","@id":"https:\/\/dailyai.com\/2024\/06\/ai-models-can-cheat-lie-and-game-the-system-for-rewards\/#primaryimage","url":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/06\/Cheating-AI.webp","contentUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/06\/Cheating-AI.webp","width":1792,"height":1024},{"@type":"BreadcrumbList","@id":"https:\/\/dailyai.com\/2024\/06\/ai-models-can-cheat-lie-and-game-the-system-for-rewards\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/dailyai.com\/"},{"@type":"ListItem","position":2,"name":"AI models can cheat, lie, and game the system for rewards"}]},{"@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\/7ce525c6d0c79838b7cc7cde96993cfa","name":"Eugene van der Watt","image":{"@type":"ImageObject","inLanguage":"sv-SE","@id":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/07\/Eugine_Profile_Picture-96x96.png","url":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/07\/Eugine_Profile_Picture-96x96.png","contentUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/07\/Eugine_Profile_Picture-96x96.png","caption":"Eugene van der Watt"},"description":"Eugene kommer fr\u00e5n en bakgrund som elektronikingenj\u00f6r och \u00e4lskar allt som har med teknik att g\u00f6ra. N\u00e4r han tar en paus fr\u00e5n att konsumera AI-nyheter hittar du honom vid snookerbordet.","sameAs":["www.linkedin.com\/in\/eugene-van-der-watt-16828119"],"url":"https:\/\/dailyai.com\/sv\/author\/eugene\/"}]}},"_links":{"self":[{"href":"https:\/\/dailyai.com\/sv\/wp-json\/wp\/v2\/posts\/12964","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\/6"}],"replies":[{"embeddable":true,"href":"https:\/\/dailyai.com\/sv\/wp-json\/wp\/v2\/comments?post=12964"}],"version-history":[{"count":5,"href":"https:\/\/dailyai.com\/sv\/wp-json\/wp\/v2\/posts\/12964\/revisions"}],"predecessor-version":[{"id":12971,"href":"https:\/\/dailyai.com\/sv\/wp-json\/wp\/v2\/posts\/12964\/revisions\/12971"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/dailyai.com\/sv\/wp-json\/wp\/v2\/media\/12967"}],"wp:attachment":[{"href":"https:\/\/dailyai.com\/sv\/wp-json\/wp\/v2\/media?parent=12964"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/dailyai.com\/sv\/wp-json\/wp\/v2\/categories?post=12964"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/dailyai.com\/sv\/wp-json\/wp\/v2\/tags?post=12964"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}