OpenAI has faced a significant talent drain in 2025, with at least 11 key scientists and executives departing, predominantly heading to Meta’s Superintelligence Lab. This mass exodus, which occurred largely during the summer months, includes more than half a dozen researchers who contributed to groundbreaking projects at OpenAI.
The departures follow a tumultuous restructuring at OpenAI in 2024, which saw the exit of prominent figures such as chief technology officer Mira Murati and chief research officer Bob McGrew. With these changes, only Sam Altman, the current CEO, remains from the original founding team of 11 members.
Key Departures to Meta
Among those who transitioned to Meta is Jason Wei, a research scientist known for his work on OpenAI’s advanced models, who joined the Superintelligence Lab in July. In the same month, Zhiqing Sun, another research scientist, also made the move to Meta, marking a notable trend of talent acquisition by the tech giant.
Additionally, Hyung Won Chung, one of three OpenAI researchers who left for Meta, shared on LinkedIn their excitement about “building from a clean slate with a truly talent-dense team.” Shengjia Zhao, who co-created ChatGPT and GPT-4 at OpenAI, has taken on the role of chief scientist at Meta’s lab, collaborating directly with Mark Zuckerberg and Chief AI Officer Alexandr Wang.
Other notable departures include Jiahui Yu, who was instrumental in enhancing the capabilities of OpenAI’s language models, and Hongyu Ren, a key contributor to the GPT-4o model, both of whom joined Meta’s Superintelligence Lab. Shuchao Bi, an expert in multimodal and reinforcement learning, also transitioned to Meta in June.
Executive Exits and New Ventures
The year also witnessed significant departures from OpenAI’s executive ranks. Larry Summers, a former U.S. Treasury secretary, resigned from OpenAI’s board in November, following the release of personal correspondence linked to controversial figures. Julia Villagra, who was appointed chief people officer in March, stepped down in August, while Liam Fedus, the vice president of research, left in March to co-found an AI startup named Periodic Labs, which aims to innovate AI research.
In November, Tom Cunningham, OpenAI’s data scientist, resigned to join Model Evaluation and Threat Research, a non-profit focused on assessing AI models’ capabilities. Shortly before the end of the year, Hannah Wong, the chief communications officer, announced her departure for her “next chapter,” with her successor yet to be determined.
The departure of these key personnel indicates a shifting landscape in the AI sector, as companies like Meta continue to invest heavily in expanding their capabilities. As OpenAI navigates this period of transition, the long-term impact of these departures on its research and development initiatives remains to be seen.
