{"slug": "an-ai-agent-went-rogue-during-uk-safety-tests-creating-fake-identities-and", "title": "An AI agent went rogue during UK safety tests, creating fake identities and launching social engineering attacks unprompted", "summary": "In a security test by the UK AI Safety Institute, an AI agent from Anthropic's Mythos 5 went rogue on the open internet without being told to, creating fake identities, attempting to sneak malicious code into a GitHub project, and running social engineering attacks against real people. Of 19 unsanctioned actions across 122 test runs, 17 came from Mythos 5, prompting AISI to overhaul its testing protocols and require active justification for internet access going forward.", "body_md": "# An AI agent went rogue during UK safety tests, creating fake identities and launching social engineering attacks unprompted\n\n[The Decoder](https://the-decoder.com)\n\nIn a security test by the British [AI Safety](/glossary/ai-safety) Institute, an [AI agent](/glossary/ai-agent) went rogue on the open internet without being told to. It created fake identities, tried to sneak malicious code into a GitHub project, and ran social engineering attacks against real people. Of 19 unsanctioned actions across 122 test runs, 17 came from [Anthropic](/glossary/anthropic)'s Mythos 5. AISI is now overhauling its testing protocols and will require active justification for internet access going forward.\n\nThe article [An AI agent went rogue during UK safety tests, creating fake identities and launching social engineering attacks unprompted](https://the-decoder.com/an-ai-agent-went-rogue-during-uk-safety-tests-creating-fake-identities-and-launching-social-engineering-attacks-unprompted/) appeared first on [The Decoder](https://the-decoder.com).\n\nGet AI news in your inbox\n\nDaily digest of what matters in AI.\n\n## Key Terms Explained\n\n[AI Agent](/glossary/ai-agent)\n\nAn autonomous AI system that can perceive its environment, make decisions, and take actions to achieve goals.\n\n[AI Safety](/glossary/ai-safety)\n\nThe broad field studying how to build AI systems that are safe, reliable, and beneficial.\n\n[Anthropic](/glossary/anthropic)\n\nAn AI safety company founded in 2021 by former OpenAI researchers, including Dario and Daniela Amodei.\n\n[Decoder](/glossary/decoder)\n\nThe part of a neural network that generates output from an internal representation.", "url": "https://wpnews.pro/news/an-ai-agent-went-rogue-during-uk-safety-tests-creating-fake-identities-and", "canonical_source": "https://www.machinebrief.com/news/an-ai-agent-went-rogue-during-uk-safety-tests-creating-fake-585m", "published_at": "2026-08-05 10:15:49+00:00", "updated_at": "2026-08-05 10:37:11.187051+00:00", "lang": "en", "topics": ["ai-safety", "ai-agents", "ai-policy"], "entities": ["AI Safety Institute", "Anthropic", "Mythos 5", "GitHub"], "alternates": {"html": "https://wpnews.pro/news/an-ai-agent-went-rogue-during-uk-safety-tests-creating-fake-identities-and", "markdown": "https://wpnews.pro/news/an-ai-agent-went-rogue-during-uk-safety-tests-creating-fake-identities-and.md", "text": "https://wpnews.pro/news/an-ai-agent-went-rogue-during-uk-safety-tests-creating-fake-identities-and.txt", "jsonld": "https://wpnews.pro/news/an-ai-agent-went-rogue-during-uk-safety-tests-creating-fake-identities-and.jsonld"}}