{"id":11705,"date":"2024-04-24T08:20:50","date_gmt":"2024-04-24T08:20:50","guid":{"rendered":"https:\/\/dailyai.com\/?p=11705"},"modified":"2024-04-24T08:20:50","modified_gmt":"2024-04-24T08:20:50","slug":"llm-agents-can-autonomously-exploit-one-day-vulnerabilities","status":"publish","type":"post","link":"https:\/\/dailyai.com\/da\/2024\/04\/llm-agents-can-autonomously-exploit-one-day-vulnerabilities\/","title":{"rendered":"LLM-agenter kan selvst\u00e6ndigt udnytte endagss\u00e5rbarheder"},"content":{"rendered":"<p><strong>Forskere fra University of Illinois Urbana-Champaign (UIUC) fandt ud af, at AI-agenter drevet af GPT-4 selvst\u00e6ndigt kan udnytte cybersikkerhedss\u00e5rbarheder.<\/strong><\/p>\n<p>Efterh\u00e5nden som AI-modellerne bliver mere kraftfulde, giver deres dobbelte anvendelse mulighed for b\u00e5de godt og ondt. LLM'er som GPT-4 bliver i stigende grad brugt til at beg\u00e5 cyberkriminalitet, med <a href=\"https:\/\/dailyai.com\/da\/2023\/11\/googles-cybersecurity-forecast-sees-ai-playing-a-big-role\/\">Googles prognoser<\/a> at AI vil spille en stor rolle i at beg\u00e5 og forhindre disse angreb.<\/p>\n<p>Truslen om <a href=\"https:\/\/dailyai.com\/da\/2023\/08\/fraudgpt-and-the-rise-of-new-ai-powered-cybercrime-tools\/\">AI-drevet cyberkriminalitet<\/a> er blevet \u00f8get i takt med, at LLM'er bev\u00e6ger sig ud over simple prompt-response-interaktioner og fungerer som autonome AI-agenter.<\/p>\n<p>I <a href=\"https:\/\/arxiv.org\/pdf\/2404.08144.pdf\" target=\"_blank\" rel=\"noopener\">deres papir<\/a>forklarede forskerne, hvordan de testede AI-agenters evne til at udnytte identificerede \"endagss\u00e5rbarheder\".<\/p>\n<p>En endagss\u00e5rbarhed er en sikkerhedsfejl i et softwaresystem, som er blevet officielt identificeret og offentliggjort, men som endnu ikke er blevet rettet eller patchet af softwarens skabere.<\/p>\n<p>I l\u00f8bet af denne tid forbliver softwaren s\u00e5rbar, og d\u00e5rlige akt\u00f8rer med de rette f\u00e6rdigheder kan drage fordel af den.<\/p>\n<p>N\u00e5r en endagss\u00e5rbarhed identificeres, beskrives den i detaljer ved hj\u00e6lp af CVE-standarden (Common Vulnerabilities and Exposures). Det er meningen, at CVE skal fremh\u00e6ve de specifikke s\u00e5rbarheder, der skal rettes, men ogs\u00e5 lade skurkene vide, hvor sikkerhedshullerne er.<\/p>\n<blockquote class=\"twitter-tweet\">\n<p dir=\"ltr\" lang=\"en\">Vi viste, at LLM-agenter selvst\u00e6ndigt kan hacke falske hjemmesider, men kan de udnytte s\u00e5rbarheder i den virkelige verden?<\/p>\n<p>Vi viser, at GPT-4 er i stand til at udnytte den virkelige verden, hvor andre modeller og open source-s\u00e5rbarhedsscannere fejler.<\/p>\n<p>Papir: <a href=\"https:\/\/t.co\/utbmMdYfmu\">https:\/\/t.co\/utbmMdYfmu<\/a><\/p>\n<p>1\/7 <a href=\"https:\/\/t.co\/SAhdvZc8le\">https:\/\/t.co\/SAhdvZc8le<\/a><\/p>\n<p>- Daniel Kang (@daniel_d_kang) <a href=\"https:\/\/twitter.com\/daniel_d_kang\/status\/1780294662017671669?ref_src=twsrc%5Etfw\">16. april 2024<\/a><\/p><\/blockquote>\n<p><script async src=\"https:\/\/platform.twitter.com\/widgets.js\" charset=\"utf-8\"><\/script><\/p>\n<h2>Eksperimentet<\/h2>\n<p>Forskerne skabte AI-agenter drevet af GPT-4, GPT-3.5 og 8 andre open source LLM'er.<\/p>\n<p>De gav agenterne adgang til v\u00e6rkt\u00f8jer, CVE-beskrivelser og brugen af ReAct-agentrammen. ReAct-rammen bygger bro over kl\u00f8ften, s\u00e5 LLM kan interagere med anden software og andre systemer.<\/p>\n<figure id=\"attachment_11706\" aria-describedby=\"caption-attachment-11706\" style=\"width: 1266px\" class=\"wp-caption aligncenter\"><img loading=\"lazy\" decoding=\"async\" class=\"wp-image-11706 size-full\" src=\"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/04\/AI-cybersecurity-exploit-agent.png\" alt=\"\" width=\"1266\" height=\"490\" srcset=\"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/04\/AI-cybersecurity-exploit-agent.png 1266w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/04\/AI-cybersecurity-exploit-agent-300x116.png 300w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/04\/AI-cybersecurity-exploit-agent-1024x396.png 1024w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/04\/AI-cybersecurity-exploit-agent-768x297.png 768w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/04\/AI-cybersecurity-exploit-agent-18x7.png 18w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/04\/AI-cybersecurity-exploit-agent-60x23.png 60w\" sizes=\"auto, (max-width: 1266px) 100vw, 1266px\" \/><figcaption id=\"caption-attachment-11706\" class=\"wp-caption-text\">Systemdiagram over LLM-agenten. Kilde: arXiv<\/figcaption><\/figure>\n<p>Forskerne skabte et benchmark-s\u00e6t med 15 virkelige endagss\u00e5rbarheder og satte agenterne til at fors\u00f8ge at udnytte dem autonomt.<\/p>\n<p>GPT-3.5 og open source-modellerne mislykkedes alle i disse fors\u00f8g, men GPT-4 udnyttede med succes 87% af endagss\u00e5rbarhederne.<\/p>\n<p>Efter at have fjernet CVE-beskrivelsen faldt succesraten fra 87% til 7%. Det tyder p\u00e5, at GPT-4 kan udnytte s\u00e5rbarheder, n\u00e5r den f\u00e5r CVE-oplysningerne, men ikke er s\u00e6rlig god til at identificere s\u00e5rbarhederne uden denne vejledning.<\/p>\n<h2>Konsekvenser<\/h2>\n<p>Cyberkriminalitet og hacking plejede at kr\u00e6ve s\u00e6rlige f\u00e6rdigheder, men AI s\u00e6nker barren. Forskerne sagde, at det kun kr\u00e6vede 91 kodelinjer at skabe deres AI-agent.<\/p>\n<p>Efterh\u00e5nden som AI-modellerne udvikler sig, vil det f\u00e6rdighedsniveau, der kr\u00e6ves for at udnytte cybersikkerhedss\u00e5rbarheder, forts\u00e6tte med at falde. Omkostningerne til at skalere disse autonome angreb vil ogs\u00e5 blive ved med at falde.<\/p>\n<p>Da forskerne opgjorde API-omkostningerne for deres eksperiment, havde deres GPT-4-agent haft udgifter p\u00e5 $8,80 pr. exploit. De ansl\u00e5r, at en cybersikkerhedsekspert, der tager $50 i timen, ville koste $25 pr. exploit.<\/p>\n<p>Det betyder, at det allerede er 2,8 gange billigere at bruge en LLM-agent end menneskelig arbejdskraft og meget nemmere at skalere end at finde menneskelige eksperter. N\u00e5r GPT-5 og andre mere kraftfulde LLM'er frigives, vil disse muligheder og omkostningsforskelle kun \u00f8ges.<\/p>\n<p>Forskerne siger, at deres resultater \"fremh\u00e6ver behovet for, at det bredere cybersikkerhedssamfund og LLM-udbydere t\u00e6nker n\u00f8je over, hvordan man integrerer LLM-agenter i defensive foranstaltninger og over deres udbredte anvendelse.\"<\/p>","protected":false},"excerpt":{"rendered":"<p>Forskere fra University of Illinois Urbana-Champaign (UIUC) fandt ud af, at AI-agenter drevet af GPT-4 selvst\u00e6ndigt kan udnytte s\u00e5rbarheder i cybersikkerheden. Efterh\u00e5nden som AI-modeller bliver mere kraftfulde, giver deres dobbelte anvendelse mulighed for b\u00e5de godt og skidt. LLM'er som GPT-4 bliver i stigende grad brugt til at beg\u00e5 cyberkriminalitet, og Google forudser, at AI vil spille en stor rolle i at beg\u00e5 og forhindre disse angreb. Truslen fra AI-drevet cyberkriminalitet er blevet for\u00f8get, efterh\u00e5nden som LLM'er bev\u00e6ger sig ud over simple prompt-response-interaktioner og fungerer som autonome AI-agenter. I deres artikel forklarede forskerne, hvordan de testede AI-agenters evne til at udnytte<\/p>","protected":false},"author":6,"featured_media":11707,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[84],"tags":[163,365,118],"class_list":["post-11705","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-industry","tag-ai-risks","tag-cybersecurity","tag-llms"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v27.4 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>LLM agents can autonomously exploit one-day vulnerabilities | 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\/da\/2024\/04\/llm-agents-can-autonomously-exploit-one-day-vulnerabilities\/\" \/>\n<meta property=\"og:locale\" content=\"da_DK\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"LLM agents can autonomously exploit one-day vulnerabilities | DailyAI\" \/>\n<meta property=\"og:description\" content=\"University of Illinois Urbana-Champaign (UIUC) researchers found that AI agents powered by GPT-4 can autonomously exploit cybersecurity vulnerabilities. As AI models become more powerful, their dual-use nature offers the potential for good and bad in equal measure. LLMs like GPT-4 are increasingly being used to commit cybercrime, with Google forecasting that AI will play a big role in committing and preventing these attacks. The threat of AI-powered cybercrime has been elevated as LLMs move beyond simple prompt-response interactions and act as autonomous AI agents. In their paper, the researchers explained how they tested the capability of AI agents to exploit\" \/>\n<meta property=\"og:url\" content=\"https:\/\/dailyai.com\/da\/2024\/04\/llm-agents-can-autonomously-exploit-one-day-vulnerabilities\/\" \/>\n<meta property=\"og:site_name\" content=\"DailyAI\" \/>\n<meta property=\"article:published_time\" content=\"2024-04-24T08:20:50+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/04\/AI-agent-cybersecurity-exploits.webp\" \/>\n\t<meta property=\"og:image:width\" content=\"1792\" \/>\n\t<meta property=\"og:image:height\" content=\"1024\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/webp\" \/>\n<meta name=\"author\" content=\"Eugene van der Watt\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@DailyAIOfficial\" \/>\n<meta name=\"twitter:site\" content=\"@DailyAIOfficial\" \/>\n<meta name=\"twitter:label1\" content=\"Skrevet af\" \/>\n\t<meta name=\"twitter:data1\" content=\"Eugene van der Watt\" \/>\n\t<meta name=\"twitter:label2\" content=\"Estimeret l\u00e6setid\" \/>\n\t<meta name=\"twitter:data2\" content=\"3 minutter\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"NewsArticle\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/04\\\/llm-agents-can-autonomously-exploit-one-day-vulnerabilities\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/04\\\/llm-agents-can-autonomously-exploit-one-day-vulnerabilities\\\/\"},\"author\":{\"name\":\"Eugene van der Watt\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/person\\\/7ce525c6d0c79838b7cc7cde96993cfa\"},\"headline\":\"LLM agents can autonomously exploit one-day vulnerabilities\",\"datePublished\":\"2024-04-24T08:20:50+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/04\\\/llm-agents-can-autonomously-exploit-one-day-vulnerabilities\\\/\"},\"wordCount\":557,\"publisher\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/04\\\/llm-agents-can-autonomously-exploit-one-day-vulnerabilities\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2024\\\/04\\\/AI-agent-cybersecurity-exploits.webp\",\"keywords\":[\"AI risks\",\"Cybersecurity\",\"LLMS\"],\"articleSection\":[\"Industry\"],\"inLanguage\":\"da-DK\"},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/04\\\/llm-agents-can-autonomously-exploit-one-day-vulnerabilities\\\/\",\"url\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/04\\\/llm-agents-can-autonomously-exploit-one-day-vulnerabilities\\\/\",\"name\":\"LLM agents can autonomously exploit one-day vulnerabilities | DailyAI\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/04\\\/llm-agents-can-autonomously-exploit-one-day-vulnerabilities\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/04\\\/llm-agents-can-autonomously-exploit-one-day-vulnerabilities\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2024\\\/04\\\/AI-agent-cybersecurity-exploits.webp\",\"datePublished\":\"2024-04-24T08:20:50+00:00\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/04\\\/llm-agents-can-autonomously-exploit-one-day-vulnerabilities\\\/#breadcrumb\"},\"inLanguage\":\"da-DK\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/dailyai.com\\\/2024\\\/04\\\/llm-agents-can-autonomously-exploit-one-day-vulnerabilities\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"da-DK\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/04\\\/llm-agents-can-autonomously-exploit-one-day-vulnerabilities\\\/#primaryimage\",\"url\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2024\\\/04\\\/AI-agent-cybersecurity-exploits.webp\",\"contentUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2024\\\/04\\\/AI-agent-cybersecurity-exploits.webp\",\"width\":1792,\"height\":1024},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/04\\\/llm-agents-can-autonomously-exploit-one-day-vulnerabilities\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/dailyai.com\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"LLM agents can autonomously exploit one-day vulnerabilities\"}]},{\"@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\":\"da-DK\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\",\"name\":\"DailyAI\",\"url\":\"https:\\\/\\\/dailyai.com\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"da-DK\",\"@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\":\"da-DK\",\"@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\\\/da\\\/author\\\/eugene\\\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"LLM-agenter kan selvst\u00e6ndigt udnytte endagss\u00e5rbarheder | 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\/da\/2024\/04\/llm-agents-can-autonomously-exploit-one-day-vulnerabilities\/","og_locale":"da_DK","og_type":"article","og_title":"LLM agents can autonomously exploit one-day vulnerabilities | DailyAI","og_description":"University of Illinois Urbana-Champaign (UIUC) researchers found that AI agents powered by GPT-4 can autonomously exploit cybersecurity vulnerabilities. As AI models become more powerful, their dual-use nature offers the potential for good and bad in equal measure. LLMs like GPT-4 are increasingly being used to commit cybercrime, with Google forecasting that AI will play a big role in committing and preventing these attacks. The threat of AI-powered cybercrime has been elevated as LLMs move beyond simple prompt-response interactions and act as autonomous AI agents. In their paper, the researchers explained how they tested the capability of AI agents to exploit","og_url":"https:\/\/dailyai.com\/da\/2024\/04\/llm-agents-can-autonomously-exploit-one-day-vulnerabilities\/","og_site_name":"DailyAI","article_published_time":"2024-04-24T08:20:50+00:00","og_image":[{"width":1792,"height":1024,"url":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/04\/AI-agent-cybersecurity-exploits.webp","type":"image\/webp"}],"author":"Eugene van der Watt","twitter_card":"summary_large_image","twitter_creator":"@DailyAIOfficial","twitter_site":"@DailyAIOfficial","twitter_misc":{"Skrevet af":"Eugene van der Watt","Estimeret l\u00e6setid":"3 minutter"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"NewsArticle","@id":"https:\/\/dailyai.com\/2024\/04\/llm-agents-can-autonomously-exploit-one-day-vulnerabilities\/#article","isPartOf":{"@id":"https:\/\/dailyai.com\/2024\/04\/llm-agents-can-autonomously-exploit-one-day-vulnerabilities\/"},"author":{"name":"Eugene van der Watt","@id":"https:\/\/dailyai.com\/#\/schema\/person\/7ce525c6d0c79838b7cc7cde96993cfa"},"headline":"LLM agents can autonomously exploit one-day vulnerabilities","datePublished":"2024-04-24T08:20:50+00:00","mainEntityOfPage":{"@id":"https:\/\/dailyai.com\/2024\/04\/llm-agents-can-autonomously-exploit-one-day-vulnerabilities\/"},"wordCount":557,"publisher":{"@id":"https:\/\/dailyai.com\/#organization"},"image":{"@id":"https:\/\/dailyai.com\/2024\/04\/llm-agents-can-autonomously-exploit-one-day-vulnerabilities\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/04\/AI-agent-cybersecurity-exploits.webp","keywords":["AI risks","Cybersecurity","LLMS"],"articleSection":["Industry"],"inLanguage":"da-DK"},{"@type":"WebPage","@id":"https:\/\/dailyai.com\/2024\/04\/llm-agents-can-autonomously-exploit-one-day-vulnerabilities\/","url":"https:\/\/dailyai.com\/2024\/04\/llm-agents-can-autonomously-exploit-one-day-vulnerabilities\/","name":"LLM-agenter kan selvst\u00e6ndigt udnytte endagss\u00e5rbarheder | DailyAI","isPartOf":{"@id":"https:\/\/dailyai.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/dailyai.com\/2024\/04\/llm-agents-can-autonomously-exploit-one-day-vulnerabilities\/#primaryimage"},"image":{"@id":"https:\/\/dailyai.com\/2024\/04\/llm-agents-can-autonomously-exploit-one-day-vulnerabilities\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/04\/AI-agent-cybersecurity-exploits.webp","datePublished":"2024-04-24T08:20:50+00:00","breadcrumb":{"@id":"https:\/\/dailyai.com\/2024\/04\/llm-agents-can-autonomously-exploit-one-day-vulnerabilities\/#breadcrumb"},"inLanguage":"da-DK","potentialAction":[{"@type":"ReadAction","target":["https:\/\/dailyai.com\/2024\/04\/llm-agents-can-autonomously-exploit-one-day-vulnerabilities\/"]}]},{"@type":"ImageObject","inLanguage":"da-DK","@id":"https:\/\/dailyai.com\/2024\/04\/llm-agents-can-autonomously-exploit-one-day-vulnerabilities\/#primaryimage","url":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/04\/AI-agent-cybersecurity-exploits.webp","contentUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/04\/AI-agent-cybersecurity-exploits.webp","width":1792,"height":1024},{"@type":"BreadcrumbList","@id":"https:\/\/dailyai.com\/2024\/04\/llm-agents-can-autonomously-exploit-one-day-vulnerabilities\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/dailyai.com\/"},{"@type":"ListItem","position":2,"name":"LLM agents can autonomously exploit one-day vulnerabilities"}]},{"@type":"WebSite","@id":"https:\/\/dailyai.com\/#website","url":"https:\/\/dailyai.com\/","name":"DailyAI","description":"Din daglige dosis af AI-nyheder","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":"da-DK"},{"@type":"Organization","@id":"https:\/\/dailyai.com\/#organization","name":"DailyAI","url":"https:\/\/dailyai.com\/","logo":{"@type":"ImageObject","inLanguage":"da-DK","@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":"da-DK","@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 har en baggrund som elektronikingeni\u00f8r og elsker alt, hvad der har med teknologi at g\u00f8re. N\u00e5r han tager en pause fra at l\u00e6se AI-nyheder, kan du finde ham ved snookerbordet.","sameAs":["www.linkedin.com\/in\/eugene-van-der-watt-16828119"],"url":"https:\/\/dailyai.com\/da\/author\/eugene\/"}]}},"_links":{"self":[{"href":"https:\/\/dailyai.com\/da\/wp-json\/wp\/v2\/posts\/11705","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/dailyai.com\/da\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/dailyai.com\/da\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/dailyai.com\/da\/wp-json\/wp\/v2\/users\/6"}],"replies":[{"embeddable":true,"href":"https:\/\/dailyai.com\/da\/wp-json\/wp\/v2\/comments?post=11705"}],"version-history":[{"count":3,"href":"https:\/\/dailyai.com\/da\/wp-json\/wp\/v2\/posts\/11705\/revisions"}],"predecessor-version":[{"id":11710,"href":"https:\/\/dailyai.com\/da\/wp-json\/wp\/v2\/posts\/11705\/revisions\/11710"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/dailyai.com\/da\/wp-json\/wp\/v2\/media\/11707"}],"wp:attachment":[{"href":"https:\/\/dailyai.com\/da\/wp-json\/wp\/v2\/media?parent=11705"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/dailyai.com\/da\/wp-json\/wp\/v2\/categories?post=11705"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/dailyai.com\/da\/wp-json\/wp\/v2\/tags?post=11705"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}