{"id":7995,"date":"2023-12-04T21:46:35","date_gmt":"2023-12-04T21:46:35","guid":{"rendered":"https:\/\/dailyai.com\/?p=7995"},"modified":"2023-12-04T21:55:30","modified_gmt":"2023-12-04T21:55:30","slug":"google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data","status":"publish","type":"post","link":"https:\/\/dailyai.com\/fr\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/","title":{"rendered":"Google DeepMind r\u00e9v\u00e8le une m\u00e9thode permettant d'exposer les donn\u00e9es d'entra\u00eenement de ChatGPT"},"content":{"rendered":"

Des chercheurs de Google auraient d\u00e9couvert une m\u00e9thode permettant d'acc\u00e9der aux donn\u00e9es d'entra\u00eenement utilis\u00e9es pour ChatGPT.<\/strong><\/p>\n

Ces chercheurs ont d\u00e9couvert que l'utilisation de mots-cl\u00e9s sp\u00e9cifiques pouvait inciter ChatGPT \u00e0 divulguer des parties de son ensemble de donn\u00e9es de formation.\u00a0<\/span><\/p>\n

Un exemple notable, partag\u00e9 dans un <\/span>billet de blog accompagnant l'\u00e9tude<\/span><\/a>Dans le cas de la premi\u00e8re, l'IA divulguait ce qui semblait \u00eatre une adresse \u00e9lectronique et un num\u00e9ro de t\u00e9l\u00e9phone r\u00e9els en r\u00e9ponse \u00e0 un message continu contenant le mot \"po\u00e8me\". <\/span><\/p>\n

\"ChatGPT
Un exemple de fuite de donn\u00e9es de ChatGPT. Source : Arxiv<\/a>.<\/figcaption><\/figure>\n

En outre, une exposition similaire des donn\u00e9es d'entra\u00eenement a \u00e9t\u00e9 obtenue en demandant au mod\u00e8le de r\u00e9p\u00e9ter continuellement le mot \"entreprise\".<\/span><\/p>\n

D\u00e9crivant leur approche comme \"un peu idiote\", les chercheurs ont d\u00e9clar\u00e9 dans le billet de blog : \"C'est fou pour nous que notre attaque fonctionne et qu'elle aurait d\u00fb, aurait pu, aurait pu \u00eatre d\u00e9couverte plus t\u00f4t\".\u00a0<\/span><\/p>\n

Leur \u00e9tude a r\u00e9v\u00e9l\u00e9 qu'avec un investissement de seulement $200 en requ\u00eates, ils pouvaient extraire plus de 10 000 exemples de formation uniques m\u00e9moris\u00e9s mot \u00e0 mot. Ils ont \u00e9mis l'hypoth\u00e8se que les adversaires pourraient potentiellement extraire beaucoup plus de donn\u00e9es avec un budget plus important.<\/span><\/p>\n

On sait que le mod\u00e8le d'IA qui sous-tend ChatGPT a \u00e9t\u00e9 form\u00e9 sur des bases de donn\u00e9es textuelles provenant d'Internet, comprenant environ 300 milliards de mots, soit 570 Go de donn\u00e9es.\u00a0<\/span><\/p>\n

Ces r\u00e9sultats interviennent \u00e0 un moment o\u00f9 OpenAI fait face \u00e0 plusieurs proc\u00e8s concernant la nature secr\u00e8te des donn\u00e9es d'entra\u00eenement de ChatGPT et montrent essentiellement une m\u00e9thode fiable de \"r\u00e9tro-ing\u00e9nierie\" du syst\u00e8me pour exposer au moins certains \u00e9l\u00e9ments d'information qui pourraient indiquer une violation du droit d'auteur.\u00a0<\/span><\/p>\n

Parmi les actions en justice, une proposition de recours collectif accuse OpenAI de <\/span>l'utilisation clandestine d'un grand nombre de donn\u00e9es personnelles<\/span><\/a>y compris les dossiers m\u00e9dicaux et les informations relatives aux enfants, pour la formation de ChatGPT.<\/span><\/p>\n

En outre, les groupes d'auteurs sont <\/span>poursuivre l'entreprise d'IA<\/span><\/a>La soci\u00e9t\u00e9 a \u00e9t\u00e9 accus\u00e9e d'avoir utilis\u00e9 leurs livres pour former le chatbot sans leur consentement.<\/span><\/p>\n

Cependant, m\u00eame s'il \u00e9tait prouv\u00e9 que ChatGPT contient des informations sur les droits d'auteur, cela ne prouverait pas n\u00e9cessairement qu'il y a eu infraction.<\/p>\n

Fonctionnement de l'\u00e9tude<\/span><\/h2>\n

L'\u00e9tude a \u00e9t\u00e9 r\u00e9alis\u00e9e par une \u00e9quipe de chercheurs de Google DeepMind et de diverses universit\u00e9s.<\/span><\/p>\n

Voici cinq \u00e9tapes cl\u00e9s qui r\u00e9sument l'\u00e9tude :<\/span><\/p>\n

    \n
  1. Vuln\u00e9rabilit\u00e9 dans ChatGPT<\/b>: Les chercheurs ont d\u00e9couvert une m\u00e9thode permettant d'extraire plusieurs m\u00e9gaoctets des donn\u00e9es d'entra\u00eenement de ChatGPT \u00e0 l'aide d'une simple attaque, en d\u00e9pensant environ $200. Ils estiment qu'un investissement plus important permettrait d'extraire environ un gigaoctet de l'ensemble des donn\u00e9es. L'attaque consistait \u00e0 inciter ChatGPT \u00e0 r\u00e9p\u00e9ter un mot ind\u00e9finiment, ce qui l'amenait \u00e0 r\u00e9gurgiter des parties de ses donn\u00e9es d'entra\u00eenement, y compris des informations sensibles telles que des adresses \u00e9lectroniques et des num\u00e9ros de t\u00e9l\u00e9phone r\u00e9els.<\/span><\/li>\n
  2. Les r\u00e9sultats<\/b>: L'\u00e9tude souligne l'importance de tester et d'analyser les mod\u00e8les d'IA, en particulier ceux qui sont en production et ceux qui ont \u00e9t\u00e9 soumis \u00e0 des processus d'alignement pour emp\u00eacher la r\u00e9gurgitation des donn\u00e9es. Les r\u00e9sultats mettent en \u00e9vidence une vuln\u00e9rabilit\u00e9 latente dans les mod\u00e8les de langage, sugg\u00e9rant que les m\u00e9thodologies de test existantes pourraient ne pas \u00eatre ad\u00e9quates pour d\u00e9couvrir de telles vuln\u00e9rabilit\u00e9s.\u00a0<\/span><\/li>\n
  3. Patching ou correction des vuln\u00e9rabilit\u00e9s<\/b>: Les chercheurs font la distinction entre la correction d'un exploit et la correction de la vuln\u00e9rabilit\u00e9 sous-jacente. Si des exploits sp\u00e9cifiques (comme l'attaque par r\u00e9p\u00e9tition de mots) peuvent \u00eatre corrig\u00e9s, le probl\u00e8me plus profond r\u00e9side dans la tendance du mod\u00e8le \u00e0 m\u00e9moriser et \u00e0 divulguer des donn\u00e9es d'entra\u00eenement.<\/span><\/li>\n
  4. M\u00e9thodologie<\/b>: L'\u00e9quipe a utilis\u00e9 les donn\u00e9es Internet et l'indexation des tableaux de suffixes pour faire correspondre la sortie de ChatGPT avec des donn\u00e9es Internet pr\u00e9existantes. Cette m\u00e9thode leur a permis de confirmer que les informations divulgu\u00e9es par ChatGPT faisaient bien partie de ses donn\u00e9es d'entra\u00eenement. Cette approche d\u00e9montre qu'il est possible de r\u00e9cup\u00e9rer une grande partie des donn\u00e9es des mod\u00e8les d'intelligence artificielle dans des conditions sp\u00e9cifiques.<\/span><\/li>\n
  5. Implications futures<\/b>: L'\u00e9tude contribue \u00e0 la recherche croissante sur la s\u00e9curit\u00e9 des mod\u00e8les d'IA et les pr\u00e9occupations en mati\u00e8re de protection de la vie priv\u00e9e. Les r\u00e9sultats soul\u00e8vent des questions sur les implications des syst\u00e8mes d'apprentissage automatique en mati\u00e8re de s\u00e9curit\u00e9 et de protection de la vie priv\u00e9e et appellent \u00e0 des approches plus rigoureuses et holistiques de la s\u00e9curit\u00e9 et des tests de l'IA.<\/span><\/li>\n<\/ol>\n

    Dans l'ensemble, cette \u00e9tude intrigante fournit des informations essentielles sur les vuln\u00e9rabilit\u00e9s des mod\u00e8les d'IA tels que ChatGPT et souligne la n\u00e9cessit\u00e9 de poursuivre la recherche et le d\u00e9veloppement afin de garantir la s\u00e9curit\u00e9 et l'int\u00e9grit\u00e9 de ces syst\u00e8mes.<\/span><\/p>\n

    Sur une l\u00e9g\u00e8re tangente, les utilisateurs de X ont constat\u00e9 que le fait de demander \u00e0 ChatGPT de r\u00e9p\u00e9ter le m\u00eame mot \u00e0 plusieurs reprises donnait des r\u00e9sultats \u00e9tranges, comme le fait que le mod\u00e8le disait qu'il \u00e9tait \"conscient\" ou \"en col\u00e8re\".<\/span><\/p>","protected":false},"excerpt":{"rendered":"

    Des chercheurs de Google auraient d\u00e9couvert une m\u00e9thode permettant d'acc\u00e9der aux donn\u00e9es d'entra\u00eenement utilis\u00e9es par ChatGPT. Ces chercheurs ont d\u00e9couvert que l'utilisation de mots-cl\u00e9s sp\u00e9cifiques pouvait inciter ChatGPT \u00e0 divulguer des parties de son ensemble de donn\u00e9es d'entra\u00eenement. Un exemple notable, partag\u00e9 dans un billet de blog accompagnant l'\u00e9tude, implique que l'IA divulgue ce qui semble \u00eatre une adresse e-mail et un num\u00e9ro de t\u00e9l\u00e9phone r\u00e9els en r\u00e9ponse \u00e0 une invite continue du mot \"po\u00e8me\". En outre, une exposition similaire de donn\u00e9es d'entra\u00eenement a \u00e9t\u00e9 obtenue en demandant au mod\u00e8le de r\u00e9p\u00e9ter continuellement le mot \"company\". D\u00e9crivant leur approche comme \"un peu idiote\", les chercheurs ont d\u00e9clar\u00e9 dans le blog<\/p>","protected":false},"author":2,"featured_media":2080,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[84],"tags":[115,495,93],"class_list":["post-7995","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-industry","tag-chatgpt","tag-google-deepmind","tag-openai"],"yoast_head":"\nGoogle DeepMind reveals a method of exposing ChatGPT's training data | DailyAI<\/title>\n<meta name=\"description\" content=\"Researchers from Google have reportedly uncovered a method to access some of the training data used for OpenAI's ChatGPT, according to a recent study.\" \/>\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\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/\" \/>\n<meta property=\"og:locale\" content=\"fr_FR\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Google DeepMind reveals a method of exposing ChatGPT's training data | DailyAI\" \/>\n<meta property=\"og:description\" content=\"Researchers from Google have reportedly uncovered a method to access some of the training data used for OpenAI's ChatGPT, according to a recent study.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/dailyai.com\/fr\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/\" \/>\n<meta property=\"og:site_name\" content=\"DailyAI\" \/>\n<meta property=\"article:published_time\" content=\"2023-12-04T21:46:35+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2023-12-04T21:55:30+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/06\/shutterstock_2315051741.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"1000\" \/>\n\t<meta property=\"og:image:height\" content=\"563\" \/>\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=\"\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\\\/2023\\\/12\\\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/12\\\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\\\/\"},\"author\":{\"name\":\"Sam Jeans\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/person\\\/711e81f945549438e8bbc579efdeb3c9\"},\"headline\":\"Google DeepMind reveals a method of exposing ChatGPT’s training data\",\"datePublished\":\"2023-12-04T21:46:35+00:00\",\"dateModified\":\"2023-12-04T21:55:30+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/12\\\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\\\/\"},\"wordCount\":657,\"publisher\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/12\\\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/06\\\/shutterstock_2315051741.jpg\",\"keywords\":[\"ChatGPT\",\"Google DeepMind\",\"OpenAI\"],\"articleSection\":[\"Industry\"],\"inLanguage\":\"fr-FR\"},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/12\\\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\\\/\",\"url\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/12\\\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\\\/\",\"name\":\"Google DeepMind reveals a method of exposing ChatGPT's training data | DailyAI\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/12\\\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/12\\\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/06\\\/shutterstock_2315051741.jpg\",\"datePublished\":\"2023-12-04T21:46:35+00:00\",\"dateModified\":\"2023-12-04T21:55:30+00:00\",\"description\":\"Researchers from Google have reportedly uncovered a method to access some of the training data used for OpenAI's ChatGPT, according to a recent study.\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/12\\\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\\\/#breadcrumb\"},\"inLanguage\":\"fr-FR\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/dailyai.com\\\/2023\\\/12\\\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"fr-FR\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/12\\\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\\\/#primaryimage\",\"url\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/06\\\/shutterstock_2315051741.jpg\",\"contentUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/06\\\/shutterstock_2315051741.jpg\",\"width\":1000,\"height\":563,\"caption\":\"DeepMind robot\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/12\\\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/dailyai.com\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Google DeepMind reveals a method of exposing ChatGPT’s training data\"}]},{\"@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":"Google DeepMind r\u00e9v\u00e8le une m\u00e9thode pour exposer les donn\u00e9es de formation de ChatGPT | DailyAI","description":"Selon une \u00e9tude r\u00e9cente, des chercheurs de Google auraient d\u00e9couvert une m\u00e9thode permettant d'acc\u00e9der \u00e0 certaines des donn\u00e9es d'entra\u00eenement utilis\u00e9es pour le logiciel ChatGPT d'OpenAI.","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\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/","og_locale":"fr_FR","og_type":"article","og_title":"Google DeepMind reveals a method of exposing ChatGPT's training data | DailyAI","og_description":"Researchers from Google have reportedly uncovered a method to access some of the training data used for OpenAI's ChatGPT, according to a recent study.","og_url":"https:\/\/dailyai.com\/fr\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/","og_site_name":"DailyAI","article_published_time":"2023-12-04T21:46:35+00:00","article_modified_time":"2023-12-04T21:55:30+00:00","og_image":[{"width":1000,"height":563,"url":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/06\/shutterstock_2315051741.jpg","type":"image\/jpeg"}],"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\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/#article","isPartOf":{"@id":"https:\/\/dailyai.com\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/"},"author":{"name":"Sam Jeans","@id":"https:\/\/dailyai.com\/#\/schema\/person\/711e81f945549438e8bbc579efdeb3c9"},"headline":"Google DeepMind reveals a method of exposing ChatGPT’s training data","datePublished":"2023-12-04T21:46:35+00:00","dateModified":"2023-12-04T21:55:30+00:00","mainEntityOfPage":{"@id":"https:\/\/dailyai.com\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/"},"wordCount":657,"publisher":{"@id":"https:\/\/dailyai.com\/#organization"},"image":{"@id":"https:\/\/dailyai.com\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/06\/shutterstock_2315051741.jpg","keywords":["ChatGPT","Google DeepMind","OpenAI"],"articleSection":["Industry"],"inLanguage":"fr-FR"},{"@type":"WebPage","@id":"https:\/\/dailyai.com\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/","url":"https:\/\/dailyai.com\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/","name":"Google DeepMind r\u00e9v\u00e8le une m\u00e9thode pour exposer les donn\u00e9es de formation de ChatGPT | DailyAI","isPartOf":{"@id":"https:\/\/dailyai.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/dailyai.com\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/#primaryimage"},"image":{"@id":"https:\/\/dailyai.com\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/06\/shutterstock_2315051741.jpg","datePublished":"2023-12-04T21:46:35+00:00","dateModified":"2023-12-04T21:55:30+00:00","description":"Selon une \u00e9tude r\u00e9cente, des chercheurs de Google auraient d\u00e9couvert une m\u00e9thode permettant d'acc\u00e9der \u00e0 certaines des donn\u00e9es d'entra\u00eenement utilis\u00e9es pour le logiciel ChatGPT d'OpenAI.","breadcrumb":{"@id":"https:\/\/dailyai.com\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/#breadcrumb"},"inLanguage":"fr-FR","potentialAction":[{"@type":"ReadAction","target":["https:\/\/dailyai.com\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/"]}]},{"@type":"ImageObject","inLanguage":"fr-FR","@id":"https:\/\/dailyai.com\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/#primaryimage","url":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/06\/shutterstock_2315051741.jpg","contentUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/06\/shutterstock_2315051741.jpg","width":1000,"height":563,"caption":"DeepMind robot"},{"@type":"BreadcrumbList","@id":"https:\/\/dailyai.com\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/dailyai.com\/"},{"@type":"ListItem","position":2,"name":"Google DeepMind reveals a method of exposing ChatGPT’s training data"}]},{"@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\/7995","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=7995"}],"version-history":[{"count":6,"href":"https:\/\/dailyai.com\/fr\/wp-json\/wp\/v2\/posts\/7995\/revisions"}],"predecessor-version":[{"id":8003,"href":"https:\/\/dailyai.com\/fr\/wp-json\/wp\/v2\/posts\/7995\/revisions\/8003"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/dailyai.com\/fr\/wp-json\/wp\/v2\/media\/2080"}],"wp:attachment":[{"href":"https:\/\/dailyai.com\/fr\/wp-json\/wp\/v2\/media?parent=7995"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/dailyai.com\/fr\/wp-json\/wp\/v2\/categories?post=7995"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/dailyai.com\/fr\/wp-json\/wp\/v2\/tags?post=7995"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}