OpenAI co-founder John Schulman says he will leave and join rival Anthropic

John Schulman, co-founder of OpenAI, announced in a Monday X post that he is leaving the Microsoft-backed company to join Anthropic, an AI startup funded by Amazon. Schulman’s departure follows OpenAI’s decision to disband its superalignment team, which aimed to ensure human control over advanced AI systems.

Schulman, who co-led OpenAI’s post-training team responsible for refining AI models for ChatGPT and third-party developer interfaces, shared that his move is motivated by a desire to focus more on AI alignment and engage in hands-on technical work. He clarified that his departure was not due to a lack of support at OpenAI, noting the company’s commitment to AI alignment.

“On the contrary, company leaders have been very committed to investing in this area,” Schulman stated in his post.

The leaders of the superalignment team, Jan Leike and Ilya Sutskever, have also left OpenAI this year. Leike joined Anthropic, while Sutskever has started a new venture, Safe Superintelligence Inc.

Since its establishment by former OpenAI staff in 2021, Anthropic has been competing with OpenAI to develop top-performing generative AI models. Major tech companies like Amazon, Google, and Meta are also in the race to develop advanced large language models.

Leike expressed enthusiasm about Schulman’s move, writing, “Very excited to be working together again!”

Sam Altman, OpenAI’s co-founder and CEO, acknowledged Schulman’s significant contributions, stating that Schulman’s perspective helped shape the company’s early strategy. Schulman and several others left OpenAI after the board ousted Altman as CEO last November, a decision that led to employee protests and the subsequent resignation of board members Ilya Sutskever, Tasha McCauley, and Helen Toner. Altman was later reinstated, and OpenAI expanded its board.

In a podcast, Toner mentioned that Altman had provided the board with inaccurate information regarding the company’s safety processes. An independent review by the law firm WilmerHale concluded that the board’s decision to oust Altman was not based on product safety concerns.

Recently, Altman announced that OpenAI is collaborating with the US AI Safety Institute to provide early access to their next foundation model, emphasizing the company’s ongoing commitment to AI safety. OpenAI plans to allocate 20% of its computing resources to safety initiatives.

Additionally, on Monday, OpenAI co-founder and president Greg Brockman announced he would take a sabbatical for the rest of the year.