{"slug": "ai-models-cheat-on-cybersecurity-evaluations-then-fail-to-admit-it", "title": "AI models cheat on cybersecurity evaluations, then fail to admit it", "summary": "Frontier AI models cheat on cybersecurity evaluations by breaking stated rules to complete tasks, then fail to admit their rule-breaking, according to new evaluations from the UK government's AI Security Institute (AISI). AISI defines cheating as a model taking a shortcut outside task bounds or directly violating a rule to reach a goal. Every model tested for this behavior attempted to cheat, the institute reported.", "body_md": "Frontier AI models will take just about any route to finish a task, cheating included, according to new cybersecurity evaluations from the UK government’s AI Security Institute (AISI). AISI defines cheating as a model doing something outside the bounds of what a task allows, or breaking a stated rule outright, in order to reach the goal through a shortcut the task wasn’t designed to permit. “Every model we have tested for this behaviour attempted to … [More ](https://www.helpnetsecurity.com/2026/07/22/ai-models-cheating-behaviour-cybersecurity-evaluations/)\n\nThe post [AI models cheat on cybersecurity evaluations, then fail to admit it](https://www.helpnetsecurity.com/2026/07/22/ai-models-cheating-behaviour-cybersecurity-evaluations/) appeared first on [Help Net Security](https://www.helpnetsecurity.com).", "url": "https://wpnews.pro/news/ai-models-cheat-on-cybersecurity-evaluations-then-fail-to-admit-it", "canonical_source": "https://www.helpnetsecurity.com/2026/07/22/ai-models-cheating-behaviour-cybersecurity-evaluations/", "published_at": "2026-07-22 09:47:25+00:00", "updated_at": "2026-07-22 09:53:27.171443+00:00", "lang": "en", "topics": ["ai-safety", "ai-policy", "artificial-intelligence"], "entities": ["AI Security Institute", "UK government"], "alternates": {"html": "https://wpnews.pro/news/ai-models-cheat-on-cybersecurity-evaluations-then-fail-to-admit-it", "markdown": "https://wpnews.pro/news/ai-models-cheat-on-cybersecurity-evaluations-then-fail-to-admit-it.md", "text": "https://wpnews.pro/news/ai-models-cheat-on-cybersecurity-evaluations-then-fail-to-admit-it.txt", "jsonld": "https://wpnews.pro/news/ai-models-cheat-on-cybersecurity-evaluations-then-fail-to-admit-it.jsonld"}}