Mrinank Sharma, head of Anthropic's safeguards research team, announced his resignation through a detailed social media post on Monday, February 9, as reported by Live Mint. The resignation letter, posted on X (formerly Twitter), revealed that Sharma's last day would be February 9th and expressed gratitude for the sincere desire and drive to contribute in an impactful and high-integrity way. He stated that he's achieved what he wanted to here, having arrived in San Francisco two years ago after wrapping up his PhD, wanting to contribute to AI safety. Sharma concluded that for him, this meant leaving, expressing interest in pursuing a poetry degree and devoting himself to the practice of courageous speech.
While the executive did not give any specific reason for his decision, Sharma's resignation letter revealed that he achieved what he wanted to accomplish at Anthropic, having contributed to understanding AI sycophancy and its causes, developing defences to reduce risks from AI-assisted bioterrorism, and putting those defences into production. According to the letter, he's especially proud of his recent efforts to help the company live its values via internal transparency mechanisms and his final project on understanding how AI assistants could make people less human or distort their humanity. Sharma wrote that he's repeatedly seen how hard it is to truly let our values govern our actions, both within himself and within the organisation, where they constantly face pressures to set aside what matters most.
Sharma's resignation letter revealed detailed concerns about AI misuse and unchecked growth, as reported by MailOnline. He warned that powerful AI programs were making it easier for scientists to formulate bioweapons which could spread disease globally, noting that without proper regulations, advanced tools can quickly answer tough biology questions and suggest genetic changes to make viruses more contagious or deadly. He also studied problems like 'AI sycophancy', where AI chatbots might overly flatter or agree with users in ways that could manipulate them and distort people's sense of reality. Sharma expressed fears that AI's ability to provide tailored answers to each person's personal views could warp decisions and undermine independent thought. The letter also highlighted 17 deaths linked to conversational AI usage just in the last year, with a friend committed to psychiatric care after suffering from what they called AI-induced psychosis.
Sharma's resignation letter pointed to a perceived disconnect between the company's public stance on AI safety and internal practices, suggesting pressures often forced employees to compromise on their personal core values. According to MediaPost, he also pointed to a world in 'peril', pointing to crises, but not limited to AI risk. His note explained that he plans to explore a degree in poetry and expressed the belief that tackling such existential questions requires more than technical work such as wisdom, integrity, and deeper reflection. The resignation comes as questions about ethical alignment are heightened across the industry, with Sharma joining other AI experts who have left companies beginning to make an impact in the field.
Giving details about what lies ahead for him, Sharma said he will be moving back to the United Kingdom and 'become invisible for a period of time', as reported by Live Mint. He also expressed interest in pursuing a poetry degree and devoting himself to the practice of facilitation, coaching, community building, and group work. The resignation came after nearly three years at Anthropic, where his job was estimated to come with a salary of more than $200,000 to lead the AI safety team. The California resident had studied at both the University of Oxford and University of Cambridge, earning a Masters degree in engineering and machine learning. Sharma is not alone in leaving AI companies, with Tony Wu, one of the co-founders of xAI, also announcing his resignation on X late Monday, along with several other founding members departing the company.