Mrinank Sharma, Anthropic’s head of AI safety since August 2023, officially stepped down on Monday, citing a series of interconnected global crises extending beyond AI and bioweapons. His resignation marks the exit of a key leader who spearheaded research on AI sycophancy and bioterrorism defenses within the startup.

Alphabet, which has contributed billions to Anthropic’s funding rounds, now faces heightened investor and regulatory scrutiny over risk governance in its AI ventures. Sharma’s departure raises concerns about Anthropic’s internal alignment on safety values and may challenge the firm’s reputation as a safety-focused competitor in the AI sector.