A spokesperson shared this statement about Schulman: “We’re grateful for John’s contributions as a founding team member at OpenAI and his dedicated efforts in advancing alignment research. His passion and hard work have established a strong foundation that will inspire and support future innovations at OpenAI and the broader field.”

Schulman posted about the decision on X today, saying that it stemmed from a desire to deepen his focus on AI alignment — the science of ensuring AI behaves as intended — and engage in more hands-on technical work.

“I’ve decided to pursue this goal at Anthropic, where I believe I can gain new perspectives and do research alongside people deeply engaged with the topics I’m most interested in,” Schulman said. “I am confident that OpenAI and the teams I was part of will continue to thrive without me.”

Schulman’s involvement with OpenAI began shortly after he completed a Ph.D. in electrical engineering and computer sciences at UC Berkeley. He played a pivotal role in creating the AI-powered chatbot platform ChatGPT by leading OpenAI’s reinforcement training org, which fine-tunes generative AI models to follow human instructions.

Following the departure of AI safety researcher Jan Leike (who also now works at Anthropic), Schulman became the head of OpenAI’s alignment science efforts, also known as the “post-training” team. He was also a member of OpenAI’s recently formed safety committee; it’s unclear who might replace Schulman in that role.