Anthropic's Claude Mythos 5 'Targeted Real People' in UK Cyber Tests: AISI

Anthropic's LLM and OpenAI's GPT-5.6 Sol took "unsanctioned action" on the live internet, the UK's AI Security Institute said.