In a notable exit for OpenAI, Lilian Weng, the company’s Vice President of Research and Safety, has announced her departure. Weng, who’s been with OpenAI for seven years, revealed in a post on X that she’s ready to “reset and explore something new.” Her last day is set for November 15, though she has yet to disclose her next move.
https://x.com/lilianweng/status/1855031273690984623
Weng’s tenure saw her lead several pivotal projects, including developing a dedicated safety team following GPT-4’s launch. Her early work included joining OpenAI’s robotics team in 2018, where she contributed to building a robot hand that famously solved a Rubik’s cube. She later moved into applied AI research and helped establish OpenAI’s safety systems unit, which now includes over 80 scientists and experts focused on creating safeguards for advanced AI.
Her departure is part of a larger trend, as many of OpenAI’s safety and research leaders have left in the last year. Among them are Ilya Sutskever and Jan Leike, co-leads of the Superalignment team, and policy researcher Miles Brundage. These exits have drawn attention to OpenAI’s balance between safety concerns and product development, with some former team members expressing worries about the potential societal risks of rapidly advancing AI technology.
Despite these challenges, OpenAI says it remains committed to maintaining rigorous safety standards. A spokesperson emphasized appreciation for Weng’s contributions and confidence that the safety team will continue its work to ensure the safe, reliable deployment of AI to users worldwide.
This departure adds to a series of recent high-profile exits, including Chief Technology Officer Mira Murati and Chief Research Officer Bob McGrew. While some former OpenAI leaders have joined competitor Anthropic, others have launched their own ventures.
