{"id":500,"date":"2025-01-31T07:46:28","date_gmt":"2025-01-31T07:46:28","guid":{"rendered":"https:\/\/familiebuckens.nl\/?p=500"},"modified":"2025-01-31T07:46:28","modified_gmt":"2025-01-31T07:46:28","slug":"new-jailbreak-techniques-expose-deepseek-llm-vulnerabilities-enabling-malicious-exploits","status":"publish","type":"post","link":"https:\/\/familiebuckens.nl\/index.php\/2025\/01\/31\/new-jailbreak-techniques-expose-deepseek-llm-vulnerabilities-enabling-malicious-exploits\/","title":{"rendered":"New Jailbreak Techniques Expose DeepSeek LLM Vulnerabilities, Enabling Malicious Exploits"},"content":{"rendered":"\n<p class=\"wp-block-paragraph\">Recent revelations have exposed critical vulnerabilities in DeepSeek\u2019s large language models (LLMs), particularly DeepSeek-R1, through advanced jailbreaking techniques.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">These exploits, including \u201cBad Likert Judge,\u201d \u201cCrescendo,\u201d and \u201cDeceptive Delight,\u201d have demonstrated the ease with which malicious actors can bypass safety measures to extract harmful outputs or generate malicious code.<\/p>\n\n\n<a class=\"wp-block-read-more\" href=\"https:\/\/familiebuckens.nl\/index.php\/2025\/01\/31\/new-jailbreak-techniques-expose-deepseek-llm-vulnerabilities-enabling-malicious-exploits\/\" target=\"_self\"><a href=\"https:\/\/cybersecuritynews.com\/new-jailbreak-techniques-expose-deepseek-llm-vulnerabilities\/\">New Jailbreak Techniques Expose DeepSeek LLM Vulnerabilities, Enabling Malicious Exploits<\/a><span class=\"screen-reader-text\">: New Jailbreak Techniques Expose DeepSeek LLM Vulnerabilities, Enabling Malicious Exploits<\/span><\/a>","protected":false},"excerpt":{"rendered":"<p>Recent revelations have exposed critical vulnerabilities in DeepSeek\u2019s large language models (LLMs), particularly DeepSeek-R1, through advanced jailbreaking techniques. These exploits, including \u201cBad Likert Judge,\u201d \u201cCrescendo,\u201d and \u201cDeceptive Delight,\u201d have demonstrated the ease with which malicious actors can bypass safety measures to extract harmful outputs or generate malicious code.<\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[17],"tags":[40],"class_list":["post-500","post","type-post","status-publish","format-standard","hentry","category-vulnerability","tag-40"],"_links":{"self":[{"href":"https:\/\/familiebuckens.nl\/index.php\/wp-json\/wp\/v2\/posts\/500","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/familiebuckens.nl\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/familiebuckens.nl\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/familiebuckens.nl\/index.php\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/familiebuckens.nl\/index.php\/wp-json\/wp\/v2\/comments?post=500"}],"version-history":[{"count":1,"href":"https:\/\/familiebuckens.nl\/index.php\/wp-json\/wp\/v2\/posts\/500\/revisions"}],"predecessor-version":[{"id":501,"href":"https:\/\/familiebuckens.nl\/index.php\/wp-json\/wp\/v2\/posts\/500\/revisions\/501"}],"wp:attachment":[{"href":"https:\/\/familiebuckens.nl\/index.php\/wp-json\/wp\/v2\/media?parent=500"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/familiebuckens.nl\/index.php\/wp-json\/wp\/v2\/categories?post=500"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/familiebuckens.nl\/index.php\/wp-json\/wp\/v2\/tags?post=500"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}