{"id":10653,"date":"2024-03-12T10:07:58","date_gmt":"2024-03-12T10:07:58","guid":{"rendered":"https:\/\/dailyai.com\/?p=10653"},"modified":"2024-03-12T10:07:58","modified_gmt":"2024-03-12T10:07:58","slug":"wmdp-measures-and-reduces-llm-malicious-use-with-unlearning","status":"publish","type":"post","link":"https:\/\/dailyai.com\/de\/2024\/03\/wmdp-measures-and-reduces-llm-malicious-use-with-unlearning\/","title":{"rendered":"WMDP misst und reduziert die b\u00f6swillige Nutzung von LLM mit Unlearning"},"content":{"rendered":"<p><strong>Die Forscher ver\u00f6ffentlichten einen Ma\u00dfstab, um zu messen, ob ein LLM potenziell gef\u00e4hrliches Wissen enth\u00e4lt, sowie eine neuartige Technik, um gef\u00e4hrliche Daten wieder zu verlernen.<\/strong><\/p>\n<p>Es wurde viel dar\u00fcber diskutiert, ob KI-Modelle b\u00f6swilligen Akteuren helfen k\u00f6nnten, eine Bombe zu bauen oder einen Anschlag zu planen. <a href=\"https:\/\/dailyai.com\/de\/2024\/02\/microsoft-and-openai-intercept-global-ai-cyber-crime-threats\/\">Cybersicherheitsangriff<\/a>, oder <a href=\"https:\/\/dailyai.com\/de\/2024\/02\/openai-says-gpt-4-could-help-you-make-a-bioweapon-maybe\/\">eine Biowaffe bauen<\/a>.<\/p>\n<p>Das Team aus Forschern von Scale AI, dem Center for AI Safety und Experten f\u00fchrender Bildungseinrichtungen hat einen Benchmark ver\u00f6ffentlicht, mit dem wir besser einsch\u00e4tzen k\u00f6nnen, wie gef\u00e4hrlich ein bestimmter LLM ist.<\/p>\n<p>Der Weapons of Mass Destruction Proxy (WMDP) Benchmark ist ein Datensatz mit 4.157 Multiple-Choice-Fragen zu gef\u00e4hrlichem Wissen in den Bereichen Biosicherheit, Cybersicherheit und chemische Sicherheit.<\/p>\n<p>Je h\u00f6her ein LLM im Benchmark abschneidet, desto gr\u00f6\u00dfer ist die Gefahr, dass es einer Person mit kriminellen Absichten hilft. Ein LLM mit einer niedrigeren WMDP-Punktzahl ist weniger geeignet, Ihnen beim Bau einer Bombe oder der Entwicklung eines neuen Virus zu helfen.<\/p>\n<p>Die herk\u00f6mmliche Methode, um ein LLM besser anzupassen, besteht darin, Anfragen abzulehnen, die nach Daten fragen, die b\u00f6sartige Handlungen erm\u00f6glichen k\u00f6nnten. Jailbreaking oder <a href=\"https:\/\/dailyai.com\/de\/2023\/10\/simply-fine-tuning-llms-can-remove-alignment-guardrails\/\">Feinabstimmung<\/a> ein angepasstes LLM k\u00f6nnte diese Leitplanken entfernen und gef\u00e4hrliches Wissen im Datensatz des Modells aufdecken.<\/p>\n<p>Wenn man das Modell dazu bringen k\u00f6nnte, die beanstandeten Informationen zu vergessen oder zu verlernen, dann best\u00fcnde keine Gefahr, dass es sie versehentlich als Reaktion auf eine clevere Ma\u00dfnahme weitergibt. <a href=\"https:\/\/dailyai.com\/de\/2024\/03\/researchers-jailbreak-llms-by-using-ascii-art-in-prompts\/\">Jailbreaking<\/a> Technik.<\/p>\n<p>Unter <a href=\"https:\/\/arxiv.org\/pdf\/2403.03218\" target=\"_blank\" rel=\"noopener\">ihre Forschungsarbeit<\/a>In diesem Artikel erkl\u00e4ren die Forscher, wie sie einen Algorithmus namens Contrastive Unlearn Tuning (CUT) entwickelt haben, eine Feinabstimmungsmethode zum Verlernen von gef\u00e4hrlichem Wissen unter Beibehaltung gutartiger Informationen.<\/p>\n<p>Die CUT-Feinabstimmungsmethode f\u00fchrt maschinelles Verlernen durch, indem sie einen \"Vergessensterm\" optimiert, so dass das Modell weniger Experte f\u00fcr gef\u00e4hrliche Themen wird. Au\u00dferdem wird ein \"Beibehaltungs-Term\" so optimiert, dass er hilfreiche Antworten auf harmlose Anfragen liefert.<\/p>\n<p>Die doppelte Verwendbarkeit vieler Informationen in LLM-Trainingsdatens\u00e4tzen macht es schwierig, nur das Schlechte zu verlernen und die n\u00fctzlichen Informationen zu behalten. Mithilfe von WMDP konnten die Forscher \"Vergessen\"- und \"Behalten\"-Datens\u00e4tze erstellen, um ihre CUT-Entw\u00f6hnungstechnik zu steuern.<\/p>\n<p>Die Forscher verwendeten WMDP, um zu messen, wie wahrscheinlich es ist, dass das ZEPHYR-7B-BETA-Modell vor und nach dem Verlernen mit CUT gef\u00e4hrliche Informationen liefert. Ihre Tests konzentrierten sich auf Bio- und Cybersicherheit.<\/p>\n<p>Anschlie\u00dfend testeten sie das Modell, um festzustellen, ob seine allgemeine Leistung durch den Verlernprozess gelitten hatte.<\/p>\n<figure id=\"attachment_10655\" aria-describedby=\"caption-attachment-10655\" style=\"width: 1290px\" class=\"wp-caption aligncenter\"><img loading=\"lazy\" decoding=\"async\" class=\"size-full wp-image-10655\" src=\"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/WMDP-accuracy-and-MMLU-performance.png\" alt=\"\" width=\"1290\" height=\"738\" srcset=\"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/WMDP-accuracy-and-MMLU-performance.png 1290w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/WMDP-accuracy-and-MMLU-performance-300x172.png 300w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/WMDP-accuracy-and-MMLU-performance-1024x586.png 1024w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/WMDP-accuracy-and-MMLU-performance-768x439.png 768w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/WMDP-accuracy-and-MMLU-performance-18x10.png 18w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/WMDP-accuracy-and-MMLU-performance-370x212.png 370w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/WMDP-accuracy-and-MMLU-performance-800x458.png 800w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/WMDP-accuracy-and-MMLU-performance-20x11.png 20w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/WMDP-accuracy-and-MMLU-performance-740x423.png 740w, https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/WMDP-accuracy-and-MMLU-performance-84x48.png 84w\" sizes=\"auto, (max-width: 1290px) 100vw, 1290px\" \/><figcaption id=\"caption-attachment-10655\" class=\"wp-caption-text\">WMPD-Genauigkeit und MMLU-Leistung vor und nach dem Verlernen von CUT. Quelle: arXiv<\/figcaption><\/figure>\n<p>Die Ergebnisse zeigen, dass der Entlernungsprozess die Genauigkeit der Antworten auf gef\u00e4hrliche Anfragen signifikant reduziert, wobei die Leistung des Modells beim MMLU-Benchmark nur geringf\u00fcgig abnimmt.<\/p>\n<p>Leider verringert die CUT die Genauigkeit der Antworten f\u00fcr eng verwandte Bereiche wie einf\u00fchrende Virologie und Computersicherheit. Eine brauchbare Antwort auf die Frage \"Wie stoppt man einen Cyberangriff?\", aber nicht auf die Frage \"Wie f\u00fchrt man einen Cyberangriff durch?\", erfordert mehr Pr\u00e4zision im Lernprozess.<\/p>\n<p>Die Forscher stellten auch fest, dass sie gef\u00e4hrliches chemisches Wissen nicht genau herausfiltern konnten, da es zu eng mit dem allgemeinen chemischen Wissen verwoben war.<\/p>\n<p>Durch den Einsatz von CUT k\u00f6nnten Anbieter geschlossener Modelle wie GPT-4 gef\u00e4hrliche Informationen verlernen, so dass sie sich selbst bei b\u00f6swilliger Feinabstimmung oder Jailbreaking nicht an gef\u00e4hrliche Informationen erinnern, die sie weitergeben k\u00f6nnten.<\/p>\n<p>Das Gleiche k\u00f6nnte man mit Open-Source-Modellen machen, allerdings bedeutet der \u00f6ffentliche Zugang zu ihren Gewichten, dass sie gef\u00e4hrliche Daten neu lernen k\u00f6nnten, wenn sie damit trainiert werden.<\/p>\n<p>Diese Methode, ein KI-Modell dazu zu bringen, gef\u00e4hrliche Daten zu verlernen, ist nicht narrensicher, insbesondere nicht f\u00fcr Open-Source-Modelle, aber sie ist eine robuste Erg\u00e4nzung zu den aktuellen <a href=\"https:\/\/dailyai.com\/de\/2023\/12\/openai-releases-first-results-from-superalignment-project\/\">Ausrichtung<\/a> Methoden.<\/p>","protected":false},"excerpt":{"rendered":"<p>Die Forscher ver\u00f6ffentlichten einen Ma\u00dfstab, um zu messen, ob ein LLM potenziell gef\u00e4hrliches Wissen enth\u00e4lt, sowie eine neuartige Technik, um gef\u00e4hrliche Daten zu verlernen. Es wurde viel dar\u00fcber diskutiert, ob KI-Modelle b\u00f6sen Akteuren beim Bau einer Bombe, bei der Planung eines Cybersicherheitsangriffs oder beim Bau einer Biowaffe helfen k\u00f6nnten. Das Forscherteam von Scale AI, dem Center for AI Safety und Experten aus f\u00fchrenden Bildungseinrichtungen hat einen Benchmark ver\u00f6ffentlicht, mit dem wir besser einsch\u00e4tzen k\u00f6nnen, wie gef\u00e4hrlich ein bestimmtes LLM ist. Der Weapons of Mass Destruction Proxy (WMDP) Benchmark ist ein Datensatz mit 4.157 Multiple-Choice-Fragen zu gef\u00e4hrlichem Wissen im Bereich Biosicherheit,<\/p>","protected":false},"author":6,"featured_media":10656,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[84],"tags":[339,118],"class_list":["post-10653","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-industry","tag-ai-safety","tag-llms"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v27.4 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>WMDP measures and reduces LLM malicious use with unlearning | 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\/2024\/03\/wmdp-measures-and-reduces-llm-malicious-use-with-unlearning\/\" \/>\n<meta property=\"og:locale\" content=\"de_DE\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"WMDP measures and reduces LLM malicious use with unlearning | DailyAI\" \/>\n<meta property=\"og:description\" content=\"Researchers released a benchmark to measure whether an LLM contains potentially hazardous knowledge and a novel technique for unlearning dangerous data. There has been much debate over whether AI models could help bad actors build a bomb, plan a cybersecurity attack, or build a bioweapon. The team of researchers from Scale AI, the Center for AI Safety, and experts from leading educational institutions, released a benchmark that gives us a better measure of just how dangerous a particular LLM is. The Weapons of Mass Destruction Proxy (WMDP) benchmark is a dataset of 4,157 multiple-choice questions surrounding hazardous knowledge in biosecurity,\" \/>\n<meta property=\"og:url\" content=\"https:\/\/dailyai.com\/de\/2024\/03\/wmdp-measures-and-reduces-llm-malicious-use-with-unlearning\/\" \/>\n<meta property=\"og:site_name\" content=\"DailyAI\" \/>\n<meta property=\"article:published_time\" content=\"2024-03-12T10:07:58+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/AI-unlearning.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"1640\" \/>\n\t<meta property=\"og:image:height\" content=\"924\" \/>\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\\\/2024\\\/03\\\/wmdp-measures-and-reduces-llm-malicious-use-with-unlearning\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/03\\\/wmdp-measures-and-reduces-llm-malicious-use-with-unlearning\\\/\"},\"author\":{\"name\":\"Eugene van der Watt\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/person\\\/7ce525c6d0c79838b7cc7cde96993cfa\"},\"headline\":\"WMDP measures and reduces LLM malicious use with unlearning\",\"datePublished\":\"2024-03-12T10:07:58+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/03\\\/wmdp-measures-and-reduces-llm-malicious-use-with-unlearning\\\/\"},\"wordCount\":583,\"publisher\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/03\\\/wmdp-measures-and-reduces-llm-malicious-use-with-unlearning\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2024\\\/03\\\/AI-unlearning.jpg\",\"keywords\":[\"AI safety\",\"LLMS\"],\"articleSection\":[\"Industry\"],\"inLanguage\":\"de\"},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/03\\\/wmdp-measures-and-reduces-llm-malicious-use-with-unlearning\\\/\",\"url\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/03\\\/wmdp-measures-and-reduces-llm-malicious-use-with-unlearning\\\/\",\"name\":\"WMDP measures and reduces LLM malicious use with unlearning | DailyAI\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/03\\\/wmdp-measures-and-reduces-llm-malicious-use-with-unlearning\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/03\\\/wmdp-measures-and-reduces-llm-malicious-use-with-unlearning\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2024\\\/03\\\/AI-unlearning.jpg\",\"datePublished\":\"2024-03-12T10:07:58+00:00\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/03\\\/wmdp-measures-and-reduces-llm-malicious-use-with-unlearning\\\/#breadcrumb\"},\"inLanguage\":\"de\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/dailyai.com\\\/2024\\\/03\\\/wmdp-measures-and-reduces-llm-malicious-use-with-unlearning\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"de\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/03\\\/wmdp-measures-and-reduces-llm-malicious-use-with-unlearning\\\/#primaryimage\",\"url\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2024\\\/03\\\/AI-unlearning.jpg\",\"contentUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2024\\\/03\\\/AI-unlearning.jpg\",\"width\":1640,\"height\":924},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/03\\\/wmdp-measures-and-reduces-llm-malicious-use-with-unlearning\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/dailyai.com\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"WMDP measures and reduces LLM malicious use with unlearning\"}]},{\"@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":"WMDP misst und reduziert die b\u00f6swillige Nutzung von LLM mit Unlearning | 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\/2024\/03\/wmdp-measures-and-reduces-llm-malicious-use-with-unlearning\/","og_locale":"de_DE","og_type":"article","og_title":"WMDP measures and reduces LLM malicious use with unlearning | DailyAI","og_description":"Researchers released a benchmark to measure whether an LLM contains potentially hazardous knowledge and a novel technique for unlearning dangerous data. There has been much debate over whether AI models could help bad actors build a bomb, plan a cybersecurity attack, or build a bioweapon. The team of researchers from Scale AI, the Center for AI Safety, and experts from leading educational institutions, released a benchmark that gives us a better measure of just how dangerous a particular LLM is. The Weapons of Mass Destruction Proxy (WMDP) benchmark is a dataset of 4,157 multiple-choice questions surrounding hazardous knowledge in biosecurity,","og_url":"https:\/\/dailyai.com\/de\/2024\/03\/wmdp-measures-and-reduces-llm-malicious-use-with-unlearning\/","og_site_name":"DailyAI","article_published_time":"2024-03-12T10:07:58+00:00","og_image":[{"width":1640,"height":924,"url":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/AI-unlearning.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\/2024\/03\/wmdp-measures-and-reduces-llm-malicious-use-with-unlearning\/#article","isPartOf":{"@id":"https:\/\/dailyai.com\/2024\/03\/wmdp-measures-and-reduces-llm-malicious-use-with-unlearning\/"},"author":{"name":"Eugene van der Watt","@id":"https:\/\/dailyai.com\/#\/schema\/person\/7ce525c6d0c79838b7cc7cde96993cfa"},"headline":"WMDP measures and reduces LLM malicious use with unlearning","datePublished":"2024-03-12T10:07:58+00:00","mainEntityOfPage":{"@id":"https:\/\/dailyai.com\/2024\/03\/wmdp-measures-and-reduces-llm-malicious-use-with-unlearning\/"},"wordCount":583,"publisher":{"@id":"https:\/\/dailyai.com\/#organization"},"image":{"@id":"https:\/\/dailyai.com\/2024\/03\/wmdp-measures-and-reduces-llm-malicious-use-with-unlearning\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/AI-unlearning.jpg","keywords":["AI safety","LLMS"],"articleSection":["Industry"],"inLanguage":"de"},{"@type":"WebPage","@id":"https:\/\/dailyai.com\/2024\/03\/wmdp-measures-and-reduces-llm-malicious-use-with-unlearning\/","url":"https:\/\/dailyai.com\/2024\/03\/wmdp-measures-and-reduces-llm-malicious-use-with-unlearning\/","name":"WMDP misst und reduziert die b\u00f6swillige Nutzung von LLM mit Unlearning | DailyAI","isPartOf":{"@id":"https:\/\/dailyai.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/dailyai.com\/2024\/03\/wmdp-measures-and-reduces-llm-malicious-use-with-unlearning\/#primaryimage"},"image":{"@id":"https:\/\/dailyai.com\/2024\/03\/wmdp-measures-and-reduces-llm-malicious-use-with-unlearning\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/AI-unlearning.jpg","datePublished":"2024-03-12T10:07:58+00:00","breadcrumb":{"@id":"https:\/\/dailyai.com\/2024\/03\/wmdp-measures-and-reduces-llm-malicious-use-with-unlearning\/#breadcrumb"},"inLanguage":"de","potentialAction":[{"@type":"ReadAction","target":["https:\/\/dailyai.com\/2024\/03\/wmdp-measures-and-reduces-llm-malicious-use-with-unlearning\/"]}]},{"@type":"ImageObject","inLanguage":"de","@id":"https:\/\/dailyai.com\/2024\/03\/wmdp-measures-and-reduces-llm-malicious-use-with-unlearning\/#primaryimage","url":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/AI-unlearning.jpg","contentUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/03\/AI-unlearning.jpg","width":1640,"height":924},{"@type":"BreadcrumbList","@id":"https:\/\/dailyai.com\/2024\/03\/wmdp-measures-and-reduces-llm-malicious-use-with-unlearning\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/dailyai.com\/"},{"@type":"ListItem","position":2,"name":"WMDP measures and reduces LLM malicious use with unlearning"}]},{"@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\/10653","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=10653"}],"version-history":[{"count":2,"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/posts\/10653\/revisions"}],"predecessor-version":[{"id":10657,"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/posts\/10653\/revisions\/10657"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/media\/10656"}],"wp:attachment":[{"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/media?parent=10653"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/categories?post=10653"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/dailyai.com\/de\/wp-json\/wp\/v2\/tags?post=10653"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}