Anthropic researcher Mrinank Sharma announced his resignation in a cryptic, poetry-laden letter warning of a world “in peril”.
For those not in the know, Sharma led the Claude chatbot maker’s Safeguards Research Team since it was formed early last year and had been at the company since 2023.
He said his work covered the causes of AI sycophancy, defences against “AI-assisted bioterrorism” and what he called “one of the first AI safety cases”.
Sharma posted that it would be his last day and uploaded a copy of the letter he shared with colleagues. The note offered few specifics, because this peril was evidently too perilous for the ears of others, but it leaned hard on the idea that internal pressures can make safety values optional.
“Throughout my time here, I’ve repeatedly seen how hard it is to truly let our values govern our actions, and employees constantly face pressures to set aside what matters most.”
He widened the gloom beyond AI, writing, “I continuously find myself reckoning with our situation. The world is in peril, and not just from AI, or bioweapons, and a whole series of interconnected crises unfolding in this very moment.”
“We appear to be approaching a threshold where our wisdom must grow in equal measure to our capacity to affect the world, lest we face the consequences,” Sharma said.
The resignation came after Anthropic’s newly released Claude Cowork model helped trigger a stock market nosedive, with fears that new plugins could upend large enterprise customers and automate white-collar roles.
Sharma’s stated next move sounded less like a new lab and more like a hard left into personal reinvention.
“I hope to explore a poetry degree and devote myself to the practice of courageous speech,” he wrote.
x
