An Artificial Intelligence (AI) safety researcher, Mrinank Sharma, has resigned from the US firm Anthropic, warning that the "world is in peril" from AI, bioweapons, and interconnected crises. He plans to return to the UK to study poetry and write.

• Mrinank Sharma led a team at Anthropic researching AI safeguards, including combating AI-assisted bioterrorism risks.
• Anthropic, known for its Claude chatbot, positions itself as safety-oriented, despite facing a $1.5 billion lawsuit settlement in 2025 for allegedly using authors' work to train AI models.
• This resignation follows another from an OpenAI researcher criticising the company's decision to deploy adverts in its chatbot, which Anthropic also recently criticised in a commercial.