John Schulman, co-founder of OpenAI, announced in a social media post on Monday that he would be leaving the Microsoft-backed company to join Anthropic, an artificial intelligence startup funded by Amazon. This decision comes shortly after OpenAI disbanded a superalignment team that was dedicated to ensuring human control over AI systems that surpass human capabilities in various tasks.
Schulman, who had been a co-leader of OpenAI’s post-training team responsible for refining AI models for the ChatGPT chatbot and developing a programming interface for third-party developers, expressed his desire to deepen his focus on AI alignment and return to hands-on technical work. Despite stating that his departure was not due to a lack of support for his work at OpenAI, Schulman emphasized that the company’s leadership had been committed to investing in the area of AI alignment.
The departure of Schulman follows the exits of Jan Leike and company co-founder Ilya Sutskever from OpenAI earlier this year. Leike joined Anthropic, while Sutskever announced his involvement in starting a new company called Safe Superintelligence Inc. These transitions have occurred since OpenAI staff members founded Anthropic in 2021, resulting in competition between the two San Francisco-based businesses to develop the most advanced generative AI models capable of producing human-like text.
Sam Altman, co-founder and CEO of OpenAI, acknowledged Schulman’s contributions to the company in shaping its early strategy. However, Schulman and others chose to leave following Altman’s removal as chief executive last November, which prompted employee protests and the resignations of board members, including Tasha McCauley and Helen Toner. Altman was eventually reinstated, and OpenAI expanded its board.
Despite the leadership changes and internal upheaval, OpenAI remains committed to advancing AI safety initiatives. Altman revealed in a recent statement that the company is collaborating with the US AI Safety Institute on an agreement to provide early access to their next foundation model for joint research on AI evaluation. OpenAI has also committed to allocating 20% of its computing resources to safety initiatives.
In a separate development, Greg Brockman, another co-founder of OpenAI and its president, declared that he would be taking a sabbatical for the remainder of the year. This decision comes amidst a period of transition and reorganization within the company following the recent departures and reinstatements of key leadership figures.
The departure of John Schulman from OpenAI marks a significant shift in the company’s leadership and direction. His decision to join Anthropic reflects a desire to focus on AI alignment and technical work, contributing to the competitive landscape of generative AI models. Despite the challenges and changes faced by OpenAI, the company remains committed to advancing AI safety and exploring collaborations with external organizations in the field of artificial intelligence.
Leave a Reply