OpenAI experienced a significant wave of departures in 2025, losing at least 11 key executives and researchers, including more than half a dozen top scientists who joined Meta’s Superintelligence Lab. The exodus marks a continuation of a turbulent period for the company, which saw major leadership changes in 2024, including the departures of chief technology officer Mira Murati, chief research officer Bob McGrew, and vice president of research Barret Zoph. As of 2025, CEO Sam Altman is one of only two remaining members of OpenAI’s original 11-person founding team, highlighting the scale of the leadership turnover. Among the most notable exits were several senior researchers who moved to Meta over the summer. Jason Wei, a research scientist instrumental in developing OpenAI’s o1 and deep research models, joined Meta’s Superintelligence Lab in July. Zhiqing Sun, another research scientist, also left for Meta in the same month. Hyung Won Chung, part of a trio of researchers who departed together, announced on LinkedIn that he and his colleagues were excited to build “from a clean slate with a truly talent-dense team.” Shengjia Zhao, a key architect behind ChatGPT and GPT-4, became chief scientist at Meta’s Superintelligence Lab in July. He now works directly with Meta CEO Mark Zuckerberg and Chief AI Officer Alexandr Wang. Jiahui Yu, who led OpenAI’s Perception team in developing multimodal capabilities like image and audio processing, also joined Meta’s Superintelligence Lab in late June. Hongyu Ren, a core contributor to the GPT-4o model, was recruited by Meta over the summer. Shuchao Bi, an expert in multimodal systems and reinforcement learning, left in June to focus on AI agents and post-training research at Meta. Other departures included Larry Summers, who stepped down from OpenAI’s board in November following the release of archived emails showing past communications with Jeffrey Epstein. Julia Villagra resigned as chief people officer in August, having been promoted to the role just months earlier. Liam Fedus, former vice president of research and post-training, left in March and co-founded Periodic Labs in September, an AI startup aiming to build an “AI scientist.” Tom Cunningham, a data scientist and economic researcher, resigned in November and joined Model Evaluation and Threat Research, a nonprofit focused on assessing AI model safety and capabilities. Hannah Wong, OpenAI’s chief communications officer, announced her departure in late December, stating she was moving on to her next chapter. She will be succeeded on an interim basis by Lindsey Held Bolton while the company searches for a permanent replacement. Her next role remains undisclosed.