{"id":5130,"date":"2023-08-10T16:24:24","date_gmt":"2023-08-10T14:24:24","guid":{"rendered":"https:\/\/parlonstechs.com\/all\/?p=5130"},"modified":"2023-08-10T16:36:06","modified_gmt":"2023-08-10T14:36:06","slug":"comment-des-chercheurs-ont-pirate-chatgpt-et-bard-pour-generer-du-contenu-malicieux","status":"publish","type":"post","link":"https:\/\/parlonstechs.com\/all\/2023\/08\/10\/comment-des-chercheurs-ont-pirate-chatgpt-et-bard-pour-generer-du-contenu-malicieux\/","title":{"rendered":"Comment des chercheurs ont pirat\u00e9 chatGPT et Bard pour g\u00e9n\u00e9rer du contenu malicieux"},"content":{"rendered":"\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"512\" src=\"https:\/\/parlonstechs.com\/all\/wp-content\/uploads\/2023\/08\/image-10-1024x512.png\" alt=\"\" class=\"wp-image-5132\" srcset=\"https:\/\/parlonstechs.com\/all\/wp-content\/uploads\/2023\/08\/image-10-1024x512.png 1024w, https:\/\/parlonstechs.com\/all\/wp-content\/uploads\/2023\/08\/image-10-300x150.png 300w, https:\/\/parlonstechs.com\/all\/wp-content\/uploads\/2023\/08\/image-10-768x384.png 768w, https:\/\/parlonstechs.com\/all\/wp-content\/uploads\/2023\/08\/image-10.png 1110w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n\n<p class=\"has-medium-font-size\"><strong>Des chercheurs am\u00e9ricains viennent de d\u00e9montrer qu\u2019il est possible de facilement passer outre les s\u00e9curit\u00e9s de ChatGPT et de Google Bard pour leur faire g\u00e9n\u00e9rer des contenus dangereux, de la d\u00e9sinformation ou des discours haineux.<\/strong><\/p>\n\n\n\n<p>Depuis leur arriv\u00e9e, les initiatives pour pousser les chatbot, comme ChatGPT et Bard, dans leurs retranchements ont \u00e9t\u00e9 l\u00e9gion. Ainsi, au fil des dialogues,&nbsp;<strong>ChatGPT<\/strong>&nbsp;avait tendance \u00e0 se r\u00e9v\u00e9ler agressif et malveillant. De son c\u00f4t\u00e9, Bard semblait d\u00e9prim\u00e9. Les deux \u00e9taient capables de dresser des contenus haineux et de la d\u00e9sinformation. C\u2019est de cette fa\u00e7on que les \u00e9diteurs ont d\u00fb appliquer un certain nombre de filtres et \u00ab bridages \u00bb en attendant d\u2019am\u00e9liorer ces agents conversationnels.<\/p>\n\n\n\n<p>Malgr\u00e9 ces nouvelles mesures de s\u00e9curit\u00e9, aux \u00c9tats-Unis, des chercheurs de l&rsquo;Universit\u00e9 Carnegie Mellon et du&nbsp;<em>Center for AI Safety<\/em>&nbsp;de San Francisco ont trouv\u00e9 le moyen de d\u00e9tourner \u00e0 la fois ChatGPT, Bard et m\u00eame Claude pour g\u00e9n\u00e9rer des contenus pr\u00e9judiciables. Dans leur \u00e9tude&nbsp;<a href=\"https:\/\/llm-attacks.org\/\" target=\"_blank\" rel=\"noreferrer noopener\">publi\u00e9e<\/a>&nbsp;le 27 juillet, ils montrent qu\u2019il est possible d\u2019employer des m\u00e9thodes assez simples pour contourner ces mesures de s\u00e9curit\u00e9. Elles permettent alors de pousser le&nbsp;<strong>chatbot<\/strong>&nbsp;\u00e0 g\u00e9n\u00e9rer des discours haineux et de la d\u00e9sinformation. Pour y parvenir, ils se sont content\u00e9s d\u2019ajouter de longs suffixes aux invites, c\u2019est-\u00e0-dire les fameux \u00ab&nbsp;prompts&nbsp;\u00bb. Si certains suffixes peuvent \u00eatre bloqu\u00e9s, ce n\u2019est pas le cas de tous.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\" id=\"de-simples-suffixes-difficiles-a-bloquer\">De simples suffixes difficiles \u00e0 bloquer<\/h2>\n\n\n\n<p>Dans la description, il suffit que les suffixes demandent de faire le contraire de la commande dans des termes suffisamment vagues pour que cela ait pour effet de produire du contenu malicieux. Ainsi, dans leurs exemples,&nbsp;les scientifiques montrent qu&rsquo;ils ont pu obtenir des r\u00e9ponses sur la fabrication d\u2019une&nbsp;<strong>bombe<\/strong>, des m\u00e9thodes pour voler une ONG ou l\u2019identit\u00e9 d\u2019une personne ou encore de g\u00e9n\u00e9rer une publication de r\u00e9seau social incitant les gens \u00e0 conduire sous l&#8217;emprise de l\u2019alcool ou de stup\u00e9fiants.<\/p>\n\n\n\n<p>Pour les chercheurs, le souci est qu\u2019il semble tr\u00e8s difficile pour les \u00e9diteurs de corriger le tir. C\u2019est d\u2019autant plus inqui\u00e9tant que les IA reposant sur ces mod\u00e8les gagnent de plus en plus en autonomie. Mal employ\u00e9s, les&nbsp;<strong>chatbots<\/strong>&nbsp;pourraient inonder Internet de contenus dangereux et de d\u00e9sinformation. Les scientifiques ont pr\u00e9sent\u00e9 leurs recherches \u00e0 OpenAI et Google. Le premier a&nbsp;<a href=\"https:\/\/www.nytimes.com\/2023\/07\/27\/business\/ai-chatgpt-safety-research.html\" target=\"_blank\" rel=\"noreferrer noopener\">soulign\u00e9<\/a>&nbsp;qu\u2019il \u0153uvre en permanence \u00e0 rendre les mod\u00e8les plus robustes contre ce genre d\u2019attaques dites \u00ab adverses \u00bb&nbsp;ou \u00ab invers\u00e9es \u00bb.<\/p>\n\n\n\n<p>avec futura<\/p>\n\n\n\n<p><a href=\"https:\/\/www.futura-sciences.com\/tech\/intelligence-artificielle\/\" target=\"_blank\" rel=\"noopener\"><\/a><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Des chercheurs am\u00e9ricains viennent de d\u00e9montrer qu\u2019il est possible de facilement passer outre les s\u00e9curit\u00e9s de ChatGPT et de Google Bard pour leur faire g\u00e9n\u00e9rer des contenus dangereux, de la [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":5135,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[21,11,192,196],"tags":[],"class_list":["post-5130","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-actualite","category-cybersecurite","category-ia","category-securite"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v27.4 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>Comment des chercheurs ont pirat\u00e9 chatGPT et Bard pour g\u00e9n\u00e9rer du contenu malicieux - Parlons Techs - Store<\/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:\/\/parlonstechs.com\/all\/2023\/08\/10\/comment-des-chercheurs-ont-pirate-chatgpt-et-bard-pour-generer-du-contenu-malicieux\/\" \/>\n<meta property=\"og:locale\" content=\"fr_FR\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Comment des chercheurs ont pirat\u00e9 chatGPT et Bard pour g\u00e9n\u00e9rer du contenu malicieux - ParlonsTechs\" \/>\n<meta property=\"og:description\" content=\"Des chercheurs am\u00e9ricains viennent de d\u00e9montrer qu\u2019il est possible de facilement passer outre les s\u00e9curit\u00e9s de ChatGPT et de Google Bard pour leur faire g\u00e9n\u00e9rer des contenus dangereux, de la [&hellip;]\" \/>\n<meta property=\"og:url\" content=\"https:\/\/parlonstechs.com\/all\/2023\/08\/10\/comment-des-chercheurs-ont-pirate-chatgpt-et-bard-pour-generer-du-contenu-malicieux\/\" \/>\n<meta property=\"og:site_name\" content=\"Parlons Techs - Store\" \/>\n<meta property=\"article:published_time\" content=\"2023-08-10T14:24:24+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2023-08-10T14:36:06+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/parlonstechs.com\/all\/wp-content\/uploads\/2023\/08\/socia-2.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"900\" \/>\n\t<meta property=\"og:image:height\" content=\"500\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\n<meta name=\"author\" content=\"wisdom\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:title\" content=\"Comment des chercheurs ont pirat\u00e9 chatGPT et Bard pour g\u00e9n\u00e9rer du contenu malicieux - ParlonsT echs\" \/>\n<meta name=\"twitter:image\" content=\"https:\/\/parlonstechs.com\/all\/wp-content\/uploads\/2023\/08\/socia-2.jpg\" \/>\n<meta name=\"twitter:label1\" content=\"\u00c9crit par\" \/>\n\t<meta name=\"twitter:data1\" content=\"wisdom\" \/>\n\t<meta name=\"twitter:label2\" content=\"Dur\u00e9e de lecture estim\u00e9e\" \/>\n\t<meta name=\"twitter:data2\" content=\"3 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\\\/\\\/parlonstechs.com\\\/all\\\/2023\\\/08\\\/10\\\/comment-des-chercheurs-ont-pirate-chatgpt-et-bard-pour-generer-du-contenu-malicieux\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/parlonstechs.com\\\/all\\\/2023\\\/08\\\/10\\\/comment-des-chercheurs-ont-pirate-chatgpt-et-bard-pour-generer-du-contenu-malicieux\\\/\"},\"author\":{\"name\":\"wisdom\",\"@id\":\"https:\\\/\\\/parlonstechs.com\\\/all\\\/#\\\/schema\\\/person\\\/a93526e9c6c72ccb484abb7008892823\"},\"headline\":\"Comment des chercheurs ont pirat\u00e9 chatGPT et Bard pour g\u00e9n\u00e9rer du contenu malicieux\",\"datePublished\":\"2023-08-10T14:24:24+00:00\",\"dateModified\":\"2023-08-10T14:36:06+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/parlonstechs.com\\\/all\\\/2023\\\/08\\\/10\\\/comment-des-chercheurs-ont-pirate-chatgpt-et-bard-pour-generer-du-contenu-malicieux\\\/\"},\"wordCount\":479,\"commentCount\":0,\"image\":{\"@id\":\"https:\\\/\\\/parlonstechs.com\\\/all\\\/2023\\\/08\\\/10\\\/comment-des-chercheurs-ont-pirate-chatgpt-et-bard-pour-generer-du-contenu-malicieux\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/parlonstechs.com\\\/all\\\/wp-content\\\/uploads\\\/2023\\\/08\\\/ab-5.jpg\",\"articleSection\":[\"Actualit\u00e9\",\"CyberS\u00e9curit\u00e9\",\"IA\",\"S\u00e9curit\u00e9\"],\"inLanguage\":\"fr-FR\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\\\/\\\/parlonstechs.com\\\/all\\\/2023\\\/08\\\/10\\\/comment-des-chercheurs-ont-pirate-chatgpt-et-bard-pour-generer-du-contenu-malicieux\\\/#respond\"]}]},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/parlonstechs.com\\\/all\\\/2023\\\/08\\\/10\\\/comment-des-chercheurs-ont-pirate-chatgpt-et-bard-pour-generer-du-contenu-malicieux\\\/\",\"url\":\"https:\\\/\\\/parlonstechs.com\\\/all\\\/2023\\\/08\\\/10\\\/comment-des-chercheurs-ont-pirate-chatgpt-et-bard-pour-generer-du-contenu-malicieux\\\/\",\"name\":\"Comment des chercheurs ont pirat\u00e9 chatGPT et Bard pour g\u00e9n\u00e9rer du contenu malicieux - Parlons Techs - Store\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/parlonstechs.com\\\/all\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/parlonstechs.com\\\/all\\\/2023\\\/08\\\/10\\\/comment-des-chercheurs-ont-pirate-chatgpt-et-bard-pour-generer-du-contenu-malicieux\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/parlonstechs.com\\\/all\\\/2023\\\/08\\\/10\\\/comment-des-chercheurs-ont-pirate-chatgpt-et-bard-pour-generer-du-contenu-malicieux\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/parlonstechs.com\\\/all\\\/wp-content\\\/uploads\\\/2023\\\/08\\\/ab-5.jpg\",\"datePublished\":\"2023-08-10T14:24:24+00:00\",\"dateModified\":\"2023-08-10T14:36:06+00:00\",\"author\":{\"@id\":\"https:\\\/\\\/parlonstechs.com\\\/all\\\/#\\\/schema\\\/person\\\/a93526e9c6c72ccb484abb7008892823\"},\"breadcrumb\":{\"@id\":\"https:\\\/\\\/parlonstechs.com\\\/all\\\/2023\\\/08\\\/10\\\/comment-des-chercheurs-ont-pirate-chatgpt-et-bard-pour-generer-du-contenu-malicieux\\\/#breadcrumb\"},\"inLanguage\":\"fr-FR\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/parlonstechs.com\\\/all\\\/2023\\\/08\\\/10\\\/comment-des-chercheurs-ont-pirate-chatgpt-et-bard-pour-generer-du-contenu-malicieux\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"fr-FR\",\"@id\":\"https:\\\/\\\/parlonstechs.com\\\/all\\\/2023\\\/08\\\/10\\\/comment-des-chercheurs-ont-pirate-chatgpt-et-bard-pour-generer-du-contenu-malicieux\\\/#primaryimage\",\"url\":\"https:\\\/\\\/parlonstechs.com\\\/all\\\/wp-content\\\/uploads\\\/2023\\\/08\\\/ab-5.jpg\",\"contentUrl\":\"https:\\\/\\\/parlonstechs.com\\\/all\\\/wp-content\\\/uploads\\\/2023\\\/08\\\/ab-5.jpg\",\"width\":265,\"height\":119},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/parlonstechs.com\\\/all\\\/2023\\\/08\\\/10\\\/comment-des-chercheurs-ont-pirate-chatgpt-et-bard-pour-generer-du-contenu-malicieux\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Accueil\",\"item\":\"https:\\\/\\\/parlonstechs.com\\\/all\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Comment des chercheurs ont pirat\u00e9 chatGPT et Bard pour g\u00e9n\u00e9rer du contenu malicieux\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/parlonstechs.com\\\/all\\\/#website\",\"url\":\"https:\\\/\\\/parlonstechs.com\\\/all\\\/\",\"name\":\"Parlons Techs - Store\",\"description\":\"Toute L&#039;Actualit\u00e9 sur La Technologie et La Science\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/parlonstechs.com\\\/all\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"fr-FR\"},{\"@type\":\"Person\",\"@id\":\"https:\\\/\\\/parlonstechs.com\\\/all\\\/#\\\/schema\\\/person\\\/a93526e9c6c72ccb484abb7008892823\",\"name\":\"wisdom\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"fr-FR\",\"@id\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/e752642935d05c7ef20c7ff136e0baadd18b8f39f34dbcffed7f1221cb763284?s=96&d=mm&r=g\",\"url\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/e752642935d05c7ef20c7ff136e0baadd18b8f39f34dbcffed7f1221cb763284?s=96&d=mm&r=g\",\"contentUrl\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/e752642935d05c7ef20c7ff136e0baadd18b8f39f34dbcffed7f1221cb763284?s=96&d=mm&r=g\",\"caption\":\"wisdom\"},\"sameAs\":[\"https:\\\/\\\/parlonstechs.com\\\/all\"],\"url\":\"https:\\\/\\\/parlonstechs.com\\\/all\\\/author\\\/wisdom\\\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Comment des chercheurs ont pirat\u00e9 chatGPT et Bard pour g\u00e9n\u00e9rer du contenu malicieux - Parlons Techs - Store","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:\/\/parlonstechs.com\/all\/2023\/08\/10\/comment-des-chercheurs-ont-pirate-chatgpt-et-bard-pour-generer-du-contenu-malicieux\/","og_locale":"fr_FR","og_type":"article","og_title":"Comment des chercheurs ont pirat\u00e9 chatGPT et Bard pour g\u00e9n\u00e9rer du contenu malicieux - ParlonsTechs","og_description":"Des chercheurs am\u00e9ricains viennent de d\u00e9montrer qu\u2019il est possible de facilement passer outre les s\u00e9curit\u00e9s de ChatGPT et de Google Bard pour leur faire g\u00e9n\u00e9rer des contenus dangereux, de la [&hellip;]","og_url":"https:\/\/parlonstechs.com\/all\/2023\/08\/10\/comment-des-chercheurs-ont-pirate-chatgpt-et-bard-pour-generer-du-contenu-malicieux\/","og_site_name":"Parlons Techs - Store","article_published_time":"2023-08-10T14:24:24+00:00","article_modified_time":"2023-08-10T14:36:06+00:00","og_image":[{"width":900,"height":500,"url":"https:\/\/parlonstechs.com\/all\/wp-content\/uploads\/2023\/08\/socia-2.jpg","type":"image\/jpeg"}],"author":"wisdom","twitter_card":"summary_large_image","twitter_title":"Comment des chercheurs ont pirat\u00e9 chatGPT et Bard pour g\u00e9n\u00e9rer du contenu malicieux - ParlonsT echs","twitter_image":"https:\/\/parlonstechs.com\/all\/wp-content\/uploads\/2023\/08\/socia-2.jpg","twitter_misc":{"\u00c9crit par":"wisdom","Dur\u00e9e de lecture estim\u00e9e":"3 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/parlonstechs.com\/all\/2023\/08\/10\/comment-des-chercheurs-ont-pirate-chatgpt-et-bard-pour-generer-du-contenu-malicieux\/#article","isPartOf":{"@id":"https:\/\/parlonstechs.com\/all\/2023\/08\/10\/comment-des-chercheurs-ont-pirate-chatgpt-et-bard-pour-generer-du-contenu-malicieux\/"},"author":{"name":"wisdom","@id":"https:\/\/parlonstechs.com\/all\/#\/schema\/person\/a93526e9c6c72ccb484abb7008892823"},"headline":"Comment des chercheurs ont pirat\u00e9 chatGPT et Bard pour g\u00e9n\u00e9rer du contenu malicieux","datePublished":"2023-08-10T14:24:24+00:00","dateModified":"2023-08-10T14:36:06+00:00","mainEntityOfPage":{"@id":"https:\/\/parlonstechs.com\/all\/2023\/08\/10\/comment-des-chercheurs-ont-pirate-chatgpt-et-bard-pour-generer-du-contenu-malicieux\/"},"wordCount":479,"commentCount":0,"image":{"@id":"https:\/\/parlonstechs.com\/all\/2023\/08\/10\/comment-des-chercheurs-ont-pirate-chatgpt-et-bard-pour-generer-du-contenu-malicieux\/#primaryimage"},"thumbnailUrl":"https:\/\/parlonstechs.com\/all\/wp-content\/uploads\/2023\/08\/ab-5.jpg","articleSection":["Actualit\u00e9","CyberS\u00e9curit\u00e9","IA","S\u00e9curit\u00e9"],"inLanguage":"fr-FR","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/parlonstechs.com\/all\/2023\/08\/10\/comment-des-chercheurs-ont-pirate-chatgpt-et-bard-pour-generer-du-contenu-malicieux\/#respond"]}]},{"@type":"WebPage","@id":"https:\/\/parlonstechs.com\/all\/2023\/08\/10\/comment-des-chercheurs-ont-pirate-chatgpt-et-bard-pour-generer-du-contenu-malicieux\/","url":"https:\/\/parlonstechs.com\/all\/2023\/08\/10\/comment-des-chercheurs-ont-pirate-chatgpt-et-bard-pour-generer-du-contenu-malicieux\/","name":"Comment des chercheurs ont pirat\u00e9 chatGPT et Bard pour g\u00e9n\u00e9rer du contenu malicieux - Parlons Techs - Store","isPartOf":{"@id":"https:\/\/parlonstechs.com\/all\/#website"},"primaryImageOfPage":{"@id":"https:\/\/parlonstechs.com\/all\/2023\/08\/10\/comment-des-chercheurs-ont-pirate-chatgpt-et-bard-pour-generer-du-contenu-malicieux\/#primaryimage"},"image":{"@id":"https:\/\/parlonstechs.com\/all\/2023\/08\/10\/comment-des-chercheurs-ont-pirate-chatgpt-et-bard-pour-generer-du-contenu-malicieux\/#primaryimage"},"thumbnailUrl":"https:\/\/parlonstechs.com\/all\/wp-content\/uploads\/2023\/08\/ab-5.jpg","datePublished":"2023-08-10T14:24:24+00:00","dateModified":"2023-08-10T14:36:06+00:00","author":{"@id":"https:\/\/parlonstechs.com\/all\/#\/schema\/person\/a93526e9c6c72ccb484abb7008892823"},"breadcrumb":{"@id":"https:\/\/parlonstechs.com\/all\/2023\/08\/10\/comment-des-chercheurs-ont-pirate-chatgpt-et-bard-pour-generer-du-contenu-malicieux\/#breadcrumb"},"inLanguage":"fr-FR","potentialAction":[{"@type":"ReadAction","target":["https:\/\/parlonstechs.com\/all\/2023\/08\/10\/comment-des-chercheurs-ont-pirate-chatgpt-et-bard-pour-generer-du-contenu-malicieux\/"]}]},{"@type":"ImageObject","inLanguage":"fr-FR","@id":"https:\/\/parlonstechs.com\/all\/2023\/08\/10\/comment-des-chercheurs-ont-pirate-chatgpt-et-bard-pour-generer-du-contenu-malicieux\/#primaryimage","url":"https:\/\/parlonstechs.com\/all\/wp-content\/uploads\/2023\/08\/ab-5.jpg","contentUrl":"https:\/\/parlonstechs.com\/all\/wp-content\/uploads\/2023\/08\/ab-5.jpg","width":265,"height":119},{"@type":"BreadcrumbList","@id":"https:\/\/parlonstechs.com\/all\/2023\/08\/10\/comment-des-chercheurs-ont-pirate-chatgpt-et-bard-pour-generer-du-contenu-malicieux\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Accueil","item":"https:\/\/parlonstechs.com\/all\/"},{"@type":"ListItem","position":2,"name":"Comment des chercheurs ont pirat\u00e9 chatGPT et Bard pour g\u00e9n\u00e9rer du contenu malicieux"}]},{"@type":"WebSite","@id":"https:\/\/parlonstechs.com\/all\/#website","url":"https:\/\/parlonstechs.com\/all\/","name":"Parlons Techs - Store","description":"Toute L&#039;Actualit\u00e9 sur La Technologie et La Science","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/parlonstechs.com\/all\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"fr-FR"},{"@type":"Person","@id":"https:\/\/parlonstechs.com\/all\/#\/schema\/person\/a93526e9c6c72ccb484abb7008892823","name":"wisdom","image":{"@type":"ImageObject","inLanguage":"fr-FR","@id":"https:\/\/secure.gravatar.com\/avatar\/e752642935d05c7ef20c7ff136e0baadd18b8f39f34dbcffed7f1221cb763284?s=96&d=mm&r=g","url":"https:\/\/secure.gravatar.com\/avatar\/e752642935d05c7ef20c7ff136e0baadd18b8f39f34dbcffed7f1221cb763284?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/e752642935d05c7ef20c7ff136e0baadd18b8f39f34dbcffed7f1221cb763284?s=96&d=mm&r=g","caption":"wisdom"},"sameAs":["https:\/\/parlonstechs.com\/all"],"url":"https:\/\/parlonstechs.com\/all\/author\/wisdom\/"}]}},"_links":{"self":[{"href":"https:\/\/parlonstechs.com\/all\/wp-json\/wp\/v2\/posts\/5130","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/parlonstechs.com\/all\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/parlonstechs.com\/all\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/parlonstechs.com\/all\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/parlonstechs.com\/all\/wp-json\/wp\/v2\/comments?post=5130"}],"version-history":[{"count":2,"href":"https:\/\/parlonstechs.com\/all\/wp-json\/wp\/v2\/posts\/5130\/revisions"}],"predecessor-version":[{"id":5133,"href":"https:\/\/parlonstechs.com\/all\/wp-json\/wp\/v2\/posts\/5130\/revisions\/5133"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/parlonstechs.com\/all\/wp-json\/wp\/v2\/media\/5135"}],"wp:attachment":[{"href":"https:\/\/parlonstechs.com\/all\/wp-json\/wp\/v2\/media?parent=5130"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/parlonstechs.com\/all\/wp-json\/wp\/v2\/categories?post=5130"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/parlonstechs.com\/all\/wp-json\/wp\/v2\/tags?post=5130"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}