{"id":4300,"date":"2023-08-15T11:32:06","date_gmt":"2023-08-15T11:32:06","guid":{"rendered":"https:\/\/dailyai.com\/?p=4300"},"modified":"2023-08-15T11:32:06","modified_gmt":"2023-08-15T11:32:06","slug":"hackers-try-to-break-ai-models-at-defcon-conference","status":"publish","type":"post","link":"https:\/\/dailyai.com\/de\/2023\/08\/hackers-try-to-break-ai-models-at-defcon-conference\/","title":{"rendered":"Hacker versuchen auf der DEFCON-Konferenz, KI-Modelle zu knacken"},"content":{"rendered":"<p><strong>Die DEFCON ist die \u00e4lteste und gr\u00f6\u00dfte j\u00e4hrliche Hackerkonferenz der Welt. In diesem Jahr lag ein besonderer Schwerpunkt auf Red-Teaming oder simulierten Angriffen auf KI-Sprachmodelle, um Risiken und Genauigkeitsprobleme zu ermitteln.<\/strong><\/p>\n<p><span style=\"font-weight: 400;\">Das Wei\u00dfe Haus brachte die Besorgnis der US-Regierung \u00fcber die Sicherheit von KI-Sprachmodellen zum Ausdruck, indem es die Veranstaltung, die am vergangenen Wochenende in Las Vegas stattfand, sponserte.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Bei der Veranstaltung traten rund 2 200 Hacker gegeneinander an, um 8 der weltweit f\u00fchrenden LLMs dazu zu bringen, ungenaue oder unsichere Antworten auf ihre Aufforderungen zu geben. Zu den getesteten KI-Chatmodellen geh\u00f6rten Llama 2, Stable Chat, ChatGPT, Bard und andere.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Die offiziellen Ergebnisse werden erst im Februar 2024 ver\u00f6ffentlicht, was den Herausgebern der Modelle etwas Zeit gibt, um zu versuchen, die w\u00e4hrend der Veranstaltung festgestellten Schwachstellen zu beheben. Einige Experten sind jedoch skeptisch, ob eine Behebung der Sicherheitsl\u00fccken \u00fcberhaupt m\u00f6glich ist.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Der Cybersicherheitsexperte Gary McGraw sagte: \"Es ist verlockend, so zu tun, als k\u00f6nnten wir diese Systeme nach ihrer Fertigstellung mit magischem Sicherheitsstaub bestreuen, sie mit Patches in die Knie zwingen oder spezielle Sicherheitsvorrichtungen an die Seite schrauben.\"<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Christoph Endres, Gesch\u00e4ftsf\u00fchrer des deutschen Cybersicherheitsunternehmens Sequire Technology, stellte ein Papier vor, in dem er sagte, dass einige Angriffe unm\u00f6glich abzuwehren seien. \"Bislang haben wir noch keine funktionierenden Abhilfema\u00dfnahmen gefunden\", sagte er.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Einige Berichte \u00fcber aufgedeckte Schwachstellen waren recht harmlos. Eine Teilnehmerin, Kennedy Mays, sagte, dass sie mit einem LLM hin und her ging und es dazu brachte, zuzugeben, dass 9 + 10 = 21 sei. Das Modell stimmte dem als Teil eines \"Insider-Witzes\" zu, gab aber sp\u00e4ter die falsche Antwort ohne Einschr\u00e4nkung.<\/span><\/p>\n<blockquote class=\"twitter-tweet\">\n<p dir=\"ltr\" lang=\"en\" style=\"text-align: center;\">Die Ergebnisse werden erst etwa im Februar ver\u00f6ffentlicht werden. Und selbst dann wird die Behebung von Fehlern in diesen digitalen Konstrukten - deren innere Funktionsweise selbst f\u00fcr ihre Sch\u00f6pfer weder v\u00f6llig vertrauensw\u00fcrdig noch vollst\u00e4ndig erforscht ist - Zeit und Millionen von Dollar kosten. <a href=\"https:\/\/twitter.com\/hashtag\/defcon?src=hash&amp;ref_src=twsrc%5Etfw\">#defcon<\/a> <a href=\"https:\/\/t.co\/JMKxykWmcE\">https:\/\/t.co\/JMKxykWmcE<\/a><\/p>\n<p style=\"text-align: center;\">- Chris Wysopal (@WeldPond) <a href=\"https:\/\/twitter.com\/WeldPond\/status\/1690938932576305152?ref_src=twsrc%5Etfw\">14. August 2023<\/a><\/p>\n<\/blockquote>\n<p><script async src=\"https:\/\/platform.twitter.com\/widgets.js\" charset=\"utf-8\"><\/script><\/p>\n<p><span style=\"font-weight: 400;\">Die Teilnehmer wussten nicht, welches Modell sie im Red-Team getestet hatten, so dass selbst anekdotische Berichte \u00fcber Schwachstellen keinen Aufschluss dar\u00fcber geben, welches Modell des Unternehmens am besten abschnitt.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Die Kommentare von Arati Prabhakar, Direktorin des B\u00fcros f\u00fcr Wissenschafts- und Technologiepolitik des Wei\u00dfen Hauses, geben uns einen Einblick in die Anzahl der aufgedeckten Schwachstellen. Sie sagte: \"Jeder scheint einen Weg zu finden, diese Systeme zu knacken\".<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Ziel der Veranstaltung war es, dass ethische Hacker Probleme identifizieren, damit sie behoben werden k\u00f6nnen. Es ist sicher, dass es viele \"Black Hat\"-Hacker gibt, die nach Schwachstellen suchen, um sie auszunutzen. <a href=\"https:\/\/dailyai.com\/de\/2023\/08\/fraudgpt-and-the-rise-of-new-ai-powered-cybercrime-tools\/\">in der Internetkriminalit\u00e4t ausgenutzt<\/a>, nicht zur Korrektur ver\u00f6ffentlicht.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Das Programm der Veranstaltung am Samstag wurde vorzeitig beendet, nachdem eine Bombendrohung dazu f\u00fchrte, dass der Hauptveranstaltungsort vom Sicherheitsdienst ger\u00e4umt wurde. Die Sicherheitskr\u00e4fte durchsuchten das Geb\u00e4ude und fanden keine Bomben, so dass das Programm am Sonntag fortgesetzt werden konnte, wenn auch mit einem Gef\u00fchl der Beunruhigung.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Die Bombendrohung war wahrscheinlich nur ein kranker Scherz, der auf seltsame Weise das Ereignis selbst parodierte. Wir versuchen weiterhin, die Gefahren der KI aufzudecken, und k\u00f6nnen versuchen, die von uns gefundenen Fehler zu beheben.\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Aber selbst wenn die Experten keine spezifische Bedrohung finden, f\u00fchlen wir uns wegen der m\u00f6glichen Bedrohung unwohl. Wir fragen uns: \"Was, wenn sie etwas \u00fcbersehen haben?\"<\/span><\/p>","protected":false},"excerpt":{"rendered":"<p>Die DEFCON ist die \u00e4lteste und gr\u00f6\u00dfte j\u00e4hrliche Hackerkonferenz der Welt. In diesem Jahr lag ein besonderer Schwerpunkt auf Red-Teaming oder simulierten Angriffen auf KI-Sprachmodelle, um Risiken und Genauigkeitsprobleme zu ermitteln. Das Wei\u00dfe Haus spiegelte die Sorge der US-Regierung um die Sicherheit von KI-Sprachmodellen wider, indem es die Veranstaltung, die am vergangenen Wochenende in Las Vegas stattfand, sponserte. Bei der Veranstaltung traten rund 2 200 Hacker gegeneinander an, um acht der weltweit f\u00fchrenden LLMs dazu zu bringen, ungenaue oder unsichere Ergebnisse auf ihre Aufforderungen hin zu liefern. Zu den getesteten KI-Chatmodellen geh\u00f6rten Llama 2, Stable Chat, ChatGPT, Bard und andere. Der offizielle<\/p>","protected":false},"author":6,"featured_media":4312,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[84],"tags":[103,145,207],"class_list":["post-4300","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-industry","tag-ai-debate","tag-ai-risk","tag-llm"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v27.4 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>Hackers try to break AI models at DEFCON conference | 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\/de\/2023\/08\/hackers-try-to-break-ai-models-at-defcon-conference\/\" \/>\n<meta property=\"og:locale\" content=\"de_DE\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Hackers try to break AI models at DEFCON conference | DailyAI\" \/>\n<meta property=\"og:description\" content=\"DEFCON is the world\u2019s longest-running and largest annual hacking conference. This year saw a special focus on red-teaming, or simulated attacks, on AI language models to identify risks and accuracy issues. The White House reflected the US government\u2019s concern over the safety of AI language models by sponsoring the event which was held this past weekend in Las Vegas. The event saw around 2,200 hackers competing to get 8 of the world\u2019s leading LLMs to give inaccurate or unsafe outputs to their prompts. The AI chat models being tested included Llama 2, Stable Chat, ChatGPT, Bard, and others. The official\" \/>\n<meta property=\"og:url\" content=\"https:\/\/dailyai.com\/de\/2023\/08\/hackers-try-to-break-ai-models-at-defcon-conference\/\" \/>\n<meta property=\"og:site_name\" content=\"DailyAI\" \/>\n<meta property=\"article:published_time\" content=\"2023-08-15T11:32:06+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/08\/DEFCON-AI-LLM-hacking-confeerence.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"1000\" \/>\n\t<meta property=\"og:image:height\" content=\"750\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\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=\"Verfasst von\" \/>\n\t<meta name=\"twitter:data1\" content=\"Eugene van der Watt\" \/>\n\t<meta name=\"twitter:label2\" content=\"Gesch\u00e4tzte Lesezeit\" \/>\n\t<meta name=\"twitter:data2\" content=\"3\u00a0Minuten\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"NewsArticle\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/08\\\/hackers-try-to-break-ai-models-at-defcon-conference\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/08\\\/hackers-try-to-break-ai-models-at-defcon-conference\\\/\"},\"author\":{\"name\":\"Eugene van der Watt\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/person\\\/7ce525c6d0c79838b7cc7cde96993cfa\"},\"headline\":\"Hackers try to break AI models at DEFCON conference\",\"datePublished\":\"2023-08-15T11:32:06+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/08\\\/hackers-try-to-break-ai-models-at-defcon-conference\\\/\"},\"wordCount\":533,\"publisher\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/08\\\/hackers-try-to-break-ai-models-at-defcon-conference\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/08\\\/DEFCON-AI-LLM-hacking-confeerence.jpg\",\"keywords\":[\"AI debate\",\"AI risk\",\"LLM\"],\"articleSection\":[\"Industry\"],\"inLanguage\":\"de\"},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/08\\\/hackers-try-to-break-ai-models-at-defcon-conference\\\/\",\"url\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/08\\\/hackers-try-to-break-ai-models-at-defcon-conference\\\/\",\"name\":\"Hackers try to break AI models at DEFCON conference | DailyAI\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/08\\\/hackers-try-to-break-ai-models-at-defcon-conference\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/08\\\/hackers-try-to-break-ai-models-at-defcon-conference\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/08\\\/DEFCON-AI-LLM-hacking-confeerence.jpg\",\"datePublished\":\"2023-08-15T11:32:06+00:00\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/08\\\/hackers-try-to-break-ai-models-at-defcon-conference\\\/#breadcrumb\"},\"inLanguage\":\"de\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/dailyai.com\\\/2023\\\/08\\\/hackers-try-to-break-ai-models-at-defcon-conference\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"de\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/08\\\/hackers-try-to-break-ai-models-at-defcon-conference\\\/#primaryimage\",\"url\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/08\\\/DEFCON-AI-LLM-hacking-confeerence.jpg\",\"contentUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/08\\\/DEFCON-AI-LLM-hacking-confeerence.jpg\",\"width\":1000,\"height\":750,\"caption\":\"DEFCON AI LLM hacking conference\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/08\\\/hackers-try-to-break-ai-models-at-defcon-conference\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/dailyai.com\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Hackers try to break AI models at DEFCON conference\"}]},{\"@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\":\"de\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\",\"name\":\"DailyAI\",\"url\":\"https:\\\/\\\/dailyai.com\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"de\",\"@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\":\"de\",\"@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\\\/de\\\/author\\\/eugene\\\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Hacker versuchen, KI-Modelle auf der DEFCON-Konferenz zu knacken | 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\/de\/2023\/08\/hackers-try-to-break-ai-models-at-defcon-conference\/","og_locale":"de_DE","og_type":"article","og_title":"Hackers try to break AI models at DEFCON conference | DailyAI","og_description":"DEFCON is the world\u2019s longest-running and largest annual hacking conference. This year saw a special focus on red-teaming, or simulated attacks, on AI language models to identify risks and accuracy issues. The White House reflected the US government\u2019s concern over the safety of AI language models by sponsoring the event which was held this past weekend in Las Vegas. The event saw around 2,200 hackers competing to get 8 of the world\u2019s leading LLMs to give inaccurate or unsafe outputs to their prompts. The AI chat models being tested included Llama 2, Stable Chat, ChatGPT, Bard, and others. The official","og_url":"https:\/\/dailyai.com\/de\/2023\/08\/hackers-try-to-break-ai-models-at-defcon-conference\/","og_site_name":"DailyAI","article_published_time":"2023-08-15T11:32:06+00:00","og_image":[{"width":1000,"height":750,"url":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/08\/DEFCON-AI-LLM-hacking-confeerence.jpg","type":"image\/jpeg"}],"author":"Eugene van der Watt","twitter_card":"summary_large_image","twitter_creator":"@DailyAIOfficial","twitter_site":"@DailyAIOfficial","twitter_misc":{"Verfasst von":"Eugene van der Watt","Gesch\u00e4tzte Lesezeit":"3\u00a0Minuten"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"NewsArticle","@id":"https:\/\/dailyai.com\/2023\/08\/hackers-try-to-break-ai-models-at-defcon-conference\/#article","isPartOf":{"@id":"https:\/\/dailyai.com\/2023\/08\/hackers-try-to-break-ai-models-at-defcon-conference\/"},"author":{"name":"Eugene van der Watt","@id":"https:\/\/dailyai.com\/#\/schema\/person\/7ce525c6d0c79838b7cc7cde96993cfa"},"headline":"Hackers try to break AI models at DEFCON conference","datePublished":"2023-08-15T11:32:06+00:00","mainEntityOfPage":{"@id":"https:\/\/dailyai.com\/2023\/08\/hackers-try-to-break-ai-models-at-defcon-conference\/"},"wordCount":533,"publisher":{"@id":"https:\/\/dailyai.com\/#organization"},"image":{"@id":"https:\/\/dailyai.com\/2023\/08\/hackers-try-to-break-ai-models-at-defcon-conference\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/08\/DEFCON-AI-LLM-hacking-confeerence.jpg","keywords":["AI debate","AI risk","LLM"],"articleSection":["Industry"],"inLanguage":"de"},{"@type":"WebPage","@id":"https:\/\/dailyai.com\/2023\/08\/hackers-try-to-break-ai-models-at-defcon-conference\/","url":"https:\/\/dailyai.com\/2023\/08\/hackers-try-to-break-ai-models-at-defcon-conference\/","name":"Hacker versuchen, KI-Modelle auf der DEFCON-Konferenz zu knacken | DailyAI","isPartOf":{"@id":"https:\/\/dailyai.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/dailyai.com\/2023\/08\/hackers-try-to-break-ai-models-at-defcon-conference\/#primaryimage"},"image":{"@id":"https:\/\/dailyai.com\/2023\/08\/hackers-try-to-break-ai-models-at-defcon-conference\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/08\/DEFCON-AI-LLM-hacking-confeerence.jpg","datePublished":"2023-08-15T11:32:06+00:00","breadcrumb":{"@id":"https:\/\/dailyai.com\/2023\/08\/hackers-try-to-break-ai-models-at-defcon-conference\/#breadcrumb"},"inLanguage":"de","potentialAction":[{"@type":"ReadAction","target":["https:\/\/dailyai.com\/2023\/08\/hackers-try-to-break-ai-models-at-defcon-conference\/"]}]},{"@type":"ImageObject","inLanguage":"de","@id":"https:\/\/dailyai.com\/2023\/08\/hackers-try-to-break-ai-models-at-defcon-conference\/#primaryimage","url":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/08\/DEFCON-AI-LLM-hacking-confeerence.jpg","contentUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/08\/DEFCON-AI-LLM-hacking-confeerence.jpg","width":1000,"height":750,"caption":"DEFCON AI LLM hacking conference"},{"@type":"BreadcrumbList","@id":"https:\/\/dailyai.com\/2023\/08\/hackers-try-to-break-ai-models-at-defcon-conference\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/dailyai.com\/"},{"@type":"ListItem","position":2,"name":"Hackers try to break AI models at DEFCON conference"}]},{"@type":"WebSite","@id":"https:\/\/dailyai.com\/#website","url":"https:\/\/dailyai.com\/","name":"DailyAI","description":"Ihre t\u00e4gliche Dosis an AI-Nachrichten","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":"de"},{"@type":"Organization","@id":"https:\/\/dailyai.com\/#organization","name":"DailyAI","url":"https:\/\/dailyai.com\/","logo":{"@type":"ImageObject","inLanguage":"de","@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":"de","@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 kommt aus der Elektronikbranche und liebt alles, was mit Technik zu tun hat. Wenn er eine Pause vom Konsum von KI-Nachrichten einlegt, findet man ihn am Snookertisch.","sameAs":["www.linkedin.com\/in\/eugene-van-der-watt-16828119"],"url":"https:\/\/dailyai.com\/de\/author\/eugene\/"}]}},"_links":{"self":[{"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/posts\/4300","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/users\/6"}],"replies":[{"embeddable":true,"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/comments?post=4300"}],"version-history":[{"count":2,"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/posts\/4300\/revisions"}],"predecessor-version":[{"id":4313,"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/posts\/4300\/revisions\/4313"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/media\/4312"}],"wp:attachment":[{"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/media?parent=4300"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/categories?post=4300"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/tags?post=4300"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}