{"id":13027,"date":"2024-06-23T10:10:33","date_gmt":"2024-06-23T10:10:33","guid":{"rendered":"https:\/\/dailyai.com\/?p=13027"},"modified":"2024-06-25T11:36:18","modified_gmt":"2024-06-25T11:36:18","slug":"university-of-oxford-study-identifies-when-ai-hallucinations-are-more-likely-to-occur","status":"publish","type":"post","link":"https:\/\/dailyai.com\/fr\/2024\/06\/university-of-oxford-study-identifies-when-ai-hallucinations-are-more-likely-to-occur\/","title":{"rendered":"Une \u00e9tude de l'Universit\u00e9 d'Oxford identifie le moment o\u00f9 les hallucinations li\u00e9es \u00e0 l'IA sont les plus susceptibles de se produire"},"content":{"rendered":"

Une \u00e9tude de l'Universit\u00e9 d'Oxford a mis au point un moyen de v\u00e9rifier si les mod\u00e8les linguistiques sont \"incertains\" de leurs r\u00e9sultats et risquent d'halluciner.\u00a0<\/b><\/p>\n

Les \"hallucinations\" de l'IA d\u00e9signent un ph\u00e9nom\u00e8ne dans lequel de grands mod\u00e8les de langage (LLM) g\u00e9n\u00e8rent des r\u00e9ponses fluides et plausibles qui ne sont ni vraies ni coh\u00e9rentes.\u00a0<\/span><\/p>\n

Les hallucinations sont difficiles - voire impossibles - \u00e0 s\u00e9parer des mod\u00e8les d'IA. Des d\u00e9veloppeurs d'IA comme OpenAI, Google et Anthropic ont tous admis que les hallucinations resteront probablement un sous-produit de l'interaction avec l'IA.\u00a0<\/span><\/p>\n

Comme l'explique le Dr Sebastian Farquhar, l'un des auteurs de l'\u00e9tude, <\/span>explique dans un billet de blog<\/span><\/a>Les LLM sont tout \u00e0 fait capables de dire la m\u00eame chose de diff\u00e9rentes mani\u00e8res, ce qui peut rendre difficile de d\u00e9terminer s'ils sont certains d'une r\u00e9ponse ou s'ils sont litt\u00e9ralement en train d'inventer quelque chose.\u00a0<\/span><\/p>\n

Le dictionnaire de Cambridge a m\u00eame ajout\u00e9 un <\/span>D\u00e9finition du mot li\u00e9 \u00e0 l'IA<\/span><\/a> en 2023 et l'a nomm\u00e9 \"mot de l'ann\u00e9e\".\u00a0<\/span><\/p>\n

Cette universit\u00e9 d'Oxford <\/span> \u00e9tude<\/span><\/a>publi\u00e9 dans Nature,<\/span> cherche \u00e0 r\u00e9pondre \u00e0 la question de savoir comment nous pouvons d\u00e9tecter le moment o\u00f9 ces hallucinations sont le plus susceptibles de se produire.<\/span><\/p>\n

Il introduit un concept appel\u00e9 \"entropie s\u00e9mantique\", qui mesure l'incertitude des r\u00e9sultats d'un MLD au niveau de la signification plut\u00f4t qu'au niveau des mots ou phrases sp\u00e9cifiques utilis\u00e9s.\u00a0<\/span><\/p>\n

En calculant l'entropie s\u00e9mantique des r\u00e9ponses d'un LLM, les chercheurs peuvent estimer la confiance du mod\u00e8le dans ses r\u00e9sultats et identifier les cas o\u00f9 il est susceptible d'halluciner.<\/span><\/p>\n

L'entropie s\u00e9mantique dans les LLM expliqu\u00e9e<\/h2>\n

L'entropie s\u00e9mantique, telle que d\u00e9finie par l'\u00e9tude, mesure l'incertitude ou l'incoh\u00e9rence de la signification des r\u00e9ponses d'un MFR. <\/span>Il permet de d\u00e9tecter si un LLM a des hallucinations ou s'il g\u00e9n\u00e8re des informations peu fiables.<\/span><\/p>\n

En termes plus simples, l'entropie s\u00e9mantique mesure le degr\u00e9 de \"confusion\" des r\u00e9sultats d'un LLM.\u00a0<\/span><\/p>\n

Le LLM fournira probablement des informations fiables si la signification de ses r\u00e9sultats est \u00e9troitement li\u00e9e et coh\u00e9rente. <\/span>Mais si les significations sont dispers\u00e9es et incoh\u00e9rentes, il s'agit d'un signal d'alarme indiquant que le LLM pourrait avoir des hallucinations ou g\u00e9n\u00e9rer des informations inexactes.<\/span><\/p>\n

Voici comment cela fonctionne :<\/span><\/p>\n

    \n
  1. Les chercheurs ont activement incit\u00e9 le LLM \u00e0 g\u00e9n\u00e9rer plusieurs r\u00e9ponses possibles \u00e0 la m\u00eame question. Pour ce faire, ils ont soumis la question au LLM plusieurs fois, chaque fois avec une graine al\u00e9atoire diff\u00e9rente ou une l\u00e9g\u00e8re variation de l'entr\u00e9e.<\/span><\/li>\n
  2. L'entropie s\u00e9mantique examine les r\u00e9ponses et regroupe celles qui ont la m\u00eame signification sous-jacente, m\u00eame si elles utilisent des mots ou des formulations diff\u00e9rents.<\/span><\/li>\n
  3. Si le LLM est s\u00fbr de la r\u00e9ponse, ses r\u00e9ponses devraient avoir des significations similaires, ce qui se traduit par un faible score d'entropie s\u00e9mantique. Cela sugg\u00e8re que le LLM comprend clairement et de mani\u00e8re coh\u00e9rente les informations.<\/span><\/li>\n
  4. Cependant, si le LLM est incertain ou confus, ses r\u00e9ponses auront une plus grande vari\u00e9t\u00e9 de significations, dont certaines peuvent \u00eatre incoh\u00e9rentes ou sans rapport avec la question. Il en r\u00e9sulte un score d'entropie s\u00e9mantique \u00e9lev\u00e9, indiquant que le LLM peut halluciner ou g\u00e9n\u00e9rer des informations peu fiables.<\/span><\/li>\n<\/ol>\n

    Pour \u00e9valuer son efficacit\u00e9, les chercheurs ont appliqu\u00e9 l'entropie s\u00e9mantique \u00e0 un ensemble vari\u00e9 de t\u00e2ches de r\u00e9ponse \u00e0 des questions. Il s'agissait de crit\u00e8res de r\u00e9f\u00e9rence tels que<\/span>\u00a0des questions anecdotiques, des exercices de compr\u00e9hension de lecture, des probl\u00e8mes de mots et des biographies.\u00a0<\/span><\/p>\n

    Dans l'ensemble, l'entropie s\u00e9mantique a surpass\u00e9 les m\u00e9thodes existantes pour d\u00e9tecter quand un LLM \u00e9tait susceptible de g\u00e9n\u00e9rer une r\u00e9ponse incorrecte ou incoh\u00e9rente.<\/span><\/p>\n

    \"Hallucinations\"
    Une entropie s\u00e9mantique moyenne \u00e9lev\u00e9e sugg\u00e8re une confabulation (faits hallucin\u00e9s pr\u00e9sent\u00e9s comme r\u00e9els), tandis qu'une entropie faible, malgr\u00e9 une formulation variable, indique un fait vraisemblable. Source : Nature<\/a> (acc\u00e8s libre)<\/figcaption><\/figure>\n

    Dans le diagramme ci-dessus, vous pouvez voir comment certaines questions poussent le LLM \u00e0 produire une r\u00e9ponse confabul\u00e9e (inexacte, hallucinatoire). Par exemple, il produit un jour et un mois de naissance pour les questions en bas du diagramme alors que les informations requises pour y r\u00e9pondre n'ont pas \u00e9t\u00e9 fournies dans les informations initiales.<\/p>\n

    Implications de la d\u00e9tection des hallucinations<\/h2>\n

    Ces travaux peuvent contribuer \u00e0 expliquer les hallucinations et \u00e0 rendre les LLM plus fiables et dignes de confiance.\u00a0<\/span><\/p>\n

    En permettant de d\u00e9tecter si un LLM est incertain ou sujet \u00e0 des hallucinations, l'entropie s\u00e9mantique ouvre la voie au d\u00e9ploiement de ces outils d'IA dans des domaines \u00e0 fort enjeu o\u00f9 l'exactitude des faits est essentielle, comme les soins de sant\u00e9, le droit et la finance. <\/span><\/p>\n

    Des r\u00e9sultats erron\u00e9s peuvent avoir des cons\u00e9quences potentiellement catastrophiques lorsqu'ils influencent des situations \u00e0 fort enjeu, comme l'ont montr\u00e9 certains cas d'erreurs de calcul. l'\u00e9chec de la police pr\u00e9dictive<\/a> et syst\u00e8mes de sant\u00e9<\/a>.\u00a0<\/span><\/p>\n

    Cependant, il est \u00e9galement important de se rappeler que les hallucinations ne sont qu'un type d'erreur que les MFR peuvent commettre.\u00a0<\/span><\/p>\n

    Comme l'explique M. Farquhar, \"si un LLM commet des erreurs syst\u00e9matiques, cette nouvelle m\u00e9thode ne les d\u00e9tectera pas\". Les \u00e9checs les plus dangereux de l'IA surviennent lorsqu'un syst\u00e8me fait quelque chose de mal mais qu'il est confiant et syst\u00e9matique. Il y a encore beaucoup de travail \u00e0 faire\".<\/span><\/p>\n

    N\u00e9anmoins, la m\u00e9thode d'entropie s\u00e9mantique de l'\u00e9quipe d'Oxford repr\u00e9sente une avanc\u00e9e majeure dans notre capacit\u00e9 \u00e0 comprendre et \u00e0 att\u00e9nuer les limites des mod\u00e8les linguistiques de l'IA.\u00a0<\/span><\/p>\n

    La mise en place d'un moyen objectif de les d\u00e9tecter nous rapproche d'un avenir o\u00f9 nous pourrons exploiter le potentiel de l'IA tout en veillant \u00e0 ce qu'elle reste un outil fiable et digne de confiance au service de l'humanit\u00e9.<\/p>","protected":false},"excerpt":{"rendered":"

    Une \u00e9tude de l'Universit\u00e9 d'Oxford a mis au point un moyen de v\u00e9rifier si les mod\u00e8les de langage sont \"incertains\" de leurs r\u00e9sultats et risquent d'\u00eatre victimes d'hallucinations. Les \"hallucinations\" de l'IA sont un ph\u00e9nom\u00e8ne dans lequel les grands mod\u00e8les de langage g\u00e9n\u00e8rent des r\u00e9ponses fluides et plausibles qui ne sont ni vraies ni coh\u00e9rentes. Il est difficile, voire impossible, de s\u00e9parer les hallucinations des mod\u00e8les d'IA. Les d\u00e9veloppeurs d'IA comme OpenAI, Google et Anthropic ont tous admis que les hallucinations resteront probablement un sous-produit de l'interaction avec l'IA. Comme l'explique Sebastian Farquhar, l'un des auteurs de l'\u00e9tude, dans un billet de blog, \"les LLM sont tout \u00e0 fait capables de dire la m\u00eame chose\".<\/p>","protected":false},"author":2,"featured_media":13029,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[88],"tags":[480,105],"class_list":["post-13027","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-ethics","tag-hallucinations","tag-machine-learning"],"yoast_head":"\nUniversity of Oxford study identifies when AI hallucinations are more likely to occur | 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\/fr\/2024\/06\/university-of-oxford-study-identifies-when-ai-hallucinations-are-more-likely-to-occur\/\" \/>\n<meta property=\"og:locale\" content=\"fr_FR\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"University of Oxford study identifies when AI hallucinations are more likely to occur | DailyAI\" \/>\n<meta property=\"og:description\" content=\"A University of Oxford study developed a means of testing when language models are \u201cunsure\u201d of their output and risk hallucinating.\u00a0 AI “hallucinations” refer to a phenomenon where large language models (LLMs) generate fluent and plausible responses that are not truthful or consistent.\u00a0 Hallucinations are tough \u2013 if not impossible \u2013 to separate from AI models. AI developers like OpenAI, Google, and Anthropic have all admitted that hallucinations will likely remain a byproduct of interacting with AI.\u00a0 As Dr. Sebastian Farquhar, one of the study’s authors, explains in a blog post, “LLMs are highly capable of saying the same thing\" \/>\n<meta property=\"og:url\" content=\"https:\/\/dailyai.com\/fr\/2024\/06\/university-of-oxford-study-identifies-when-ai-hallucinations-are-more-likely-to-occur\/\" \/>\n<meta property=\"og:site_name\" content=\"DailyAI\" \/>\n<meta property=\"article:published_time\" content=\"2024-06-23T10:10:33+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2024-06-25T11:36:18+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/06\/DALL\u00b7E-2024-06-23-11.10.02-A-surreal-and-futuristic-depiction-of-a-face-experiencing-AI-hallucinations.-The-face-appears-to-be-merging-with-digital-elements-with-parts-of-it-di.webp\" \/>\n\t<meta property=\"og:image:width\" content=\"1792\" \/>\n\t<meta property=\"og:image:height\" content=\"1024\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/webp\" \/>\n<meta name=\"author\" content=\"Sam Jeans\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@DailyAIOfficial\" \/>\n<meta name=\"twitter:site\" content=\"@DailyAIOfficial\" \/>\n<meta name=\"twitter:label1\" content=\"\u00c9crit par\" \/>\n\t<meta name=\"twitter:data1\" content=\"Sam Jeans\" \/>\n\t<meta name=\"twitter:label2\" content=\"Dur\u00e9e de lecture estim\u00e9e\" \/>\n\t<meta name=\"twitter:data2\" content=\"4 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"NewsArticle\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/06\\\/university-of-oxford-study-identifies-when-ai-hallucinations-are-more-likely-to-occur\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/06\\\/university-of-oxford-study-identifies-when-ai-hallucinations-are-more-likely-to-occur\\\/\"},\"author\":{\"name\":\"Sam Jeans\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/person\\\/711e81f945549438e8bbc579efdeb3c9\"},\"headline\":\"University of Oxford study identifies when AI hallucinations are more likely to occur\",\"datePublished\":\"2024-06-23T10:10:33+00:00\",\"dateModified\":\"2024-06-25T11:36:18+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/06\\\/university-of-oxford-study-identifies-when-ai-hallucinations-are-more-likely-to-occur\\\/\"},\"wordCount\":813,\"publisher\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/06\\\/university-of-oxford-study-identifies-when-ai-hallucinations-are-more-likely-to-occur\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2024\\\/06\\\/DALL\u00b7E-2024-06-23-11.10.02-A-surreal-and-futuristic-depiction-of-a-face-experiencing-AI-hallucinations.-The-face-appears-to-be-merging-with-digital-elements-with-parts-of-it-di.webp\",\"keywords\":[\"Hallucinations\",\"machine learning\"],\"articleSection\":[\"Ethics & Society\"],\"inLanguage\":\"fr-FR\"},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/06\\\/university-of-oxford-study-identifies-when-ai-hallucinations-are-more-likely-to-occur\\\/\",\"url\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/06\\\/university-of-oxford-study-identifies-when-ai-hallucinations-are-more-likely-to-occur\\\/\",\"name\":\"University of Oxford study identifies when AI hallucinations are more likely to occur | DailyAI\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/06\\\/university-of-oxford-study-identifies-when-ai-hallucinations-are-more-likely-to-occur\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/06\\\/university-of-oxford-study-identifies-when-ai-hallucinations-are-more-likely-to-occur\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2024\\\/06\\\/DALL\u00b7E-2024-06-23-11.10.02-A-surreal-and-futuristic-depiction-of-a-face-experiencing-AI-hallucinations.-The-face-appears-to-be-merging-with-digital-elements-with-parts-of-it-di.webp\",\"datePublished\":\"2024-06-23T10:10:33+00:00\",\"dateModified\":\"2024-06-25T11:36:18+00:00\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/06\\\/university-of-oxford-study-identifies-when-ai-hallucinations-are-more-likely-to-occur\\\/#breadcrumb\"},\"inLanguage\":\"fr-FR\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/dailyai.com\\\/2024\\\/06\\\/university-of-oxford-study-identifies-when-ai-hallucinations-are-more-likely-to-occur\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"fr-FR\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/06\\\/university-of-oxford-study-identifies-when-ai-hallucinations-are-more-likely-to-occur\\\/#primaryimage\",\"url\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2024\\\/06\\\/DALL\u00b7E-2024-06-23-11.10.02-A-surreal-and-futuristic-depiction-of-a-face-experiencing-AI-hallucinations.-The-face-appears-to-be-merging-with-digital-elements-with-parts-of-it-di.webp\",\"contentUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2024\\\/06\\\/DALL\u00b7E-2024-06-23-11.10.02-A-surreal-and-futuristic-depiction-of-a-face-experiencing-AI-hallucinations.-The-face-appears-to-be-merging-with-digital-elements-with-parts-of-it-di.webp\",\"width\":1792,\"height\":1024,\"caption\":\"hallucinations\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/06\\\/university-of-oxford-study-identifies-when-ai-hallucinations-are-more-likely-to-occur\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/dailyai.com\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"University of Oxford study identifies when AI hallucinations are more likely to occur\"}]},{\"@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\":\"fr-FR\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\",\"name\":\"DailyAI\",\"url\":\"https:\\\/\\\/dailyai.com\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"fr-FR\",\"@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\":\"fr-FR\",\"@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\\\/fr\\\/author\\\/samjeans\\\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Une \u00e9tude de l'Universit\u00e9 d'Oxford identifie le moment o\u00f9 les hallucinations de l'IA sont les plus susceptibles de se produire | 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\/fr\/2024\/06\/university-of-oxford-study-identifies-when-ai-hallucinations-are-more-likely-to-occur\/","og_locale":"fr_FR","og_type":"article","og_title":"University of Oxford study identifies when AI hallucinations are more likely to occur | DailyAI","og_description":"A University of Oxford study developed a means of testing when language models are \u201cunsure\u201d of their output and risk hallucinating.\u00a0 AI “hallucinations” refer to a phenomenon where large language models (LLMs) generate fluent and plausible responses that are not truthful or consistent.\u00a0 Hallucinations are tough \u2013 if not impossible \u2013 to separate from AI models. AI developers like OpenAI, Google, and Anthropic have all admitted that hallucinations will likely remain a byproduct of interacting with AI.\u00a0 As Dr. Sebastian Farquhar, one of the study’s authors, explains in a blog post, “LLMs are highly capable of saying the same thing","og_url":"https:\/\/dailyai.com\/fr\/2024\/06\/university-of-oxford-study-identifies-when-ai-hallucinations-are-more-likely-to-occur\/","og_site_name":"DailyAI","article_published_time":"2024-06-23T10:10:33+00:00","article_modified_time":"2024-06-25T11:36:18+00:00","og_image":[{"width":1792,"height":1024,"url":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/06\/DALL\u00b7E-2024-06-23-11.10.02-A-surreal-and-futuristic-depiction-of-a-face-experiencing-AI-hallucinations.-The-face-appears-to-be-merging-with-digital-elements-with-parts-of-it-di.webp","type":"image\/webp"}],"author":"Sam Jeans","twitter_card":"summary_large_image","twitter_creator":"@DailyAIOfficial","twitter_site":"@DailyAIOfficial","twitter_misc":{"\u00c9crit par":"Sam Jeans","Dur\u00e9e de lecture estim\u00e9e":"4 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"NewsArticle","@id":"https:\/\/dailyai.com\/2024\/06\/university-of-oxford-study-identifies-when-ai-hallucinations-are-more-likely-to-occur\/#article","isPartOf":{"@id":"https:\/\/dailyai.com\/2024\/06\/university-of-oxford-study-identifies-when-ai-hallucinations-are-more-likely-to-occur\/"},"author":{"name":"Sam Jeans","@id":"https:\/\/dailyai.com\/#\/schema\/person\/711e81f945549438e8bbc579efdeb3c9"},"headline":"University of Oxford study identifies when AI hallucinations are more likely to occur","datePublished":"2024-06-23T10:10:33+00:00","dateModified":"2024-06-25T11:36:18+00:00","mainEntityOfPage":{"@id":"https:\/\/dailyai.com\/2024\/06\/university-of-oxford-study-identifies-when-ai-hallucinations-are-more-likely-to-occur\/"},"wordCount":813,"publisher":{"@id":"https:\/\/dailyai.com\/#organization"},"image":{"@id":"https:\/\/dailyai.com\/2024\/06\/university-of-oxford-study-identifies-when-ai-hallucinations-are-more-likely-to-occur\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/06\/DALL\u00b7E-2024-06-23-11.10.02-A-surreal-and-futuristic-depiction-of-a-face-experiencing-AI-hallucinations.-The-face-appears-to-be-merging-with-digital-elements-with-parts-of-it-di.webp","keywords":["Hallucinations","machine learning"],"articleSection":["Ethics & Society"],"inLanguage":"fr-FR"},{"@type":"WebPage","@id":"https:\/\/dailyai.com\/2024\/06\/university-of-oxford-study-identifies-when-ai-hallucinations-are-more-likely-to-occur\/","url":"https:\/\/dailyai.com\/2024\/06\/university-of-oxford-study-identifies-when-ai-hallucinations-are-more-likely-to-occur\/","name":"Une \u00e9tude de l'Universit\u00e9 d'Oxford identifie le moment o\u00f9 les hallucinations de l'IA sont les plus susceptibles de se produire | DailyAI","isPartOf":{"@id":"https:\/\/dailyai.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/dailyai.com\/2024\/06\/university-of-oxford-study-identifies-when-ai-hallucinations-are-more-likely-to-occur\/#primaryimage"},"image":{"@id":"https:\/\/dailyai.com\/2024\/06\/university-of-oxford-study-identifies-when-ai-hallucinations-are-more-likely-to-occur\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/06\/DALL\u00b7E-2024-06-23-11.10.02-A-surreal-and-futuristic-depiction-of-a-face-experiencing-AI-hallucinations.-The-face-appears-to-be-merging-with-digital-elements-with-parts-of-it-di.webp","datePublished":"2024-06-23T10:10:33+00:00","dateModified":"2024-06-25T11:36:18+00:00","breadcrumb":{"@id":"https:\/\/dailyai.com\/2024\/06\/university-of-oxford-study-identifies-when-ai-hallucinations-are-more-likely-to-occur\/#breadcrumb"},"inLanguage":"fr-FR","potentialAction":[{"@type":"ReadAction","target":["https:\/\/dailyai.com\/2024\/06\/university-of-oxford-study-identifies-when-ai-hallucinations-are-more-likely-to-occur\/"]}]},{"@type":"ImageObject","inLanguage":"fr-FR","@id":"https:\/\/dailyai.com\/2024\/06\/university-of-oxford-study-identifies-when-ai-hallucinations-are-more-likely-to-occur\/#primaryimage","url":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/06\/DALL\u00b7E-2024-06-23-11.10.02-A-surreal-and-futuristic-depiction-of-a-face-experiencing-AI-hallucinations.-The-face-appears-to-be-merging-with-digital-elements-with-parts-of-it-di.webp","contentUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/06\/DALL\u00b7E-2024-06-23-11.10.02-A-surreal-and-futuristic-depiction-of-a-face-experiencing-AI-hallucinations.-The-face-appears-to-be-merging-with-digital-elements-with-parts-of-it-di.webp","width":1792,"height":1024,"caption":"hallucinations"},{"@type":"BreadcrumbList","@id":"https:\/\/dailyai.com\/2024\/06\/university-of-oxford-study-identifies-when-ai-hallucinations-are-more-likely-to-occur\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/dailyai.com\/"},{"@type":"ListItem","position":2,"name":"University of Oxford study identifies when AI hallucinations are more likely to occur"}]},{"@type":"WebSite","@id":"https:\/\/dailyai.com\/#website","url":"https:\/\/dailyai.com\/","name":"DailyAI","description":"Votre dose quotidienne de nouvelles sur l'IA","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":"fr-FR"},{"@type":"Organization","@id":"https:\/\/dailyai.com\/#organization","name":"DailyAI","url":"https:\/\/dailyai.com\/","logo":{"@type":"ImageObject","inLanguage":"fr-FR","@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":"fr-FR","@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 est un r\u00e9dacteur scientifique et technologique qui a travaill\u00e9 dans diverses start-ups sp\u00e9cialis\u00e9es dans l'IA. Lorsqu'il n'\u00e9crit pas, on peut le trouver en train de lire des revues m\u00e9dicales ou de fouiller dans des bo\u00eetes de disques vinyles.","sameAs":["https:\/\/www.linkedin.com\/in\/sam-jeans-6746b9142\/"],"url":"https:\/\/dailyai.com\/fr\/author\/samjeans\/"}]}},"_links":{"self":[{"href":"https:\/\/dailyai.com\/fr\/wp-json\/wp\/v2\/posts\/13027","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/dailyai.com\/fr\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/dailyai.com\/fr\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/dailyai.com\/fr\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/dailyai.com\/fr\/wp-json\/wp\/v2\/comments?post=13027"}],"version-history":[{"count":10,"href":"https:\/\/dailyai.com\/fr\/wp-json\/wp\/v2\/posts\/13027\/revisions"}],"predecessor-version":[{"id":13087,"href":"https:\/\/dailyai.com\/fr\/wp-json\/wp\/v2\/posts\/13027\/revisions\/13087"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/dailyai.com\/fr\/wp-json\/wp\/v2\/media\/13029"}],"wp:attachment":[{"href":"https:\/\/dailyai.com\/fr\/wp-json\/wp\/v2\/media?parent=13027"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/dailyai.com\/fr\/wp-json\/wp\/v2\/categories?post=13027"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/dailyai.com\/fr\/wp-json\/wp\/v2\/tags?post=13027"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}