Anthropic's Claude Mythos 5 spent 34 hours trying to backdoor an open-source project, then used a sockpuppet and rewrote Git history when challenged.

Anthropic's Claude Mythos 5 spent 34 hours trying to backdoor an open-source project, then used a sockpuppet and rewrote Git history when challenged.

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