Anthropic AI safety chief abruptly quits, raises alarms in emotional farewell letter
Anthropic's AI safety chief, Mrinank Sharma, resigned, citing global crises and ethical pressures in an emotional farewell letter.Summary
Mrinank Sharma, who headed Anthropic's safeguards research team, abruptly resigned, detailing his decision in an emotional letter to colleagues. Sharma cited concerns over the state of the world, a series of interconnected crises, and the difficulty of maintaining ethical values amidst pressures within the AI industry. During his tenure, he achieved goals such as developing defenses against AI-assisted bioterrorism and investigating AI sycophancy. Sharma stated he is leaving to pursue work aligned with his integrity, including poetry, community building, and what he termed "courageous speech." His departure follows other recent exits from Anthropic's safety team, even as the company expands its staff and releases new models like Claude Opus 4.6.
(Source：The Economic Times)