tech
February 12, 2026
AI safety researcher quits with a cryptic warning
“The world is in peril,” Anthropic’s Safeguards Research Team lead Mrinank Sharma wrote in his resignation letter
TL;DR
- Mrinank Sharma, lead of Anthropic's Safeguards Research Team, has resigned.
- Sharma cited "interconnected crises" including AI, bioweapons, and broader societal issues as reasons for his departure.
- He expressed challenges in prioritizing safety values within Anthropic and society.
- Sharma's team focused on AI security threats, bioterrorism, and catastrophe prevention.
- He plans to move back to the UK to study poetry and become "invisible."
- Anthropic CEO Dario Amodei has also warned of the potential dangers of advanced AI, including "autonomy risks" and "global totalitarian dictatorship."
Continue reading
the original article