Another OpenAI co-founder decides to quit; plans to join Amazon, Google-backed Anthropic

OpenAI co-founder John Schulman announced on Monday via X that he is leaving the Microsoft-backed company to join Anthropic, an AI startup funded by Amazon and Google. This decision follows less than three months after OpenAI disbanded its superalignment team, which focused on ensuring AI systems remain under human control even as they surpass human capabilities in various tasks. Ilya Sutskever, another OpenAI founder, announced his departure from ChatGPT in May. 

Schulman played a crucial role at OpenAI as a co-leader of the post-training team, which fine-tuned AI models for the ChatGPT chatbot and developed a programming interface for third-party developers. In June, OpenAI announced that Schulman, as head of alignment science, would join a safety and security committee to provide advice to the board. Schulman has been with OpenAI since 2016, after earning his Ph.D. in computer science from the University of California, Berkeley.

In his social media post, Schulman explained that his decision to move was driven by a desire to focus more on AI alignment and to return to hands-on technical work. He clarified that his departure was not due to a lack of support for AI alignment at OpenAI. "On the contrary, company leaders have been very committed to investing in this area," he said.

I shared the following note with my OpenAI colleagues today:

I've made the difficult decision to leave OpenAI. This choice stems from my desire to deepen my focus on AI alignment, and to start a new chapter of my career where I can return to hands-on technical work. I've decided…

— John Schulman (@johnschulman2) August 6, 2024

Other key departures

Earlier this year, Jan Leike and co-founder Ilya Sutskever, leaders of the disbanded superalignment team, also left OpenAI. Leike joined Anthropic, and Sutskever is now involved in starting his own AI company under the name Safe Superintelligence Inc. Leike expressed enthusiasm about working with Schulman again, responding to his post with, "Very excited to be working together again!"

OpenAI co-founder and CEO Sam Altman acknowledged Schulman's contributions, noting that Schulman's perspective was integral to the startup's early strategy.

The departures come after a turbulent period at OpenAI. In November, the board ousted Altman as CEO, leading to employee protests and the subsequent resignation of board members Ilya Sutskever, Tasha McCauley, and Helen Toner. Altman was eventually reinstated, and the company added new board members. Toner later revealed in a podcast that the board's decision to remove Altman was based on misinformation about the company's safety processes.

Despite these upheavals, OpenAI continues to prioritise AI safety. Last week, Altman announced on X that OpenAI is collaborating with the US AI Safety Institute to provide early access to their next foundation model, aiming to advance AI evaluation science. Altman reiterated OpenAI's commitment to dedicating 20 per cent of its computing resources to safety initiatives.

On Monday, Greg Brockman, another co-founder and the president of OpenAI, announced that he would be taking a sabbatical for the rest of the year.