Unlock the Editor’s Digest for free
Roula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.
John Schulman, one of OpenAI’s co-founders and a key architect of its ChatGPT chatbot, has left to join its main rival Anthropic, becoming the latest in a succession of senior figures to depart the leading artificial intelligence company in recent months.
Schulman is leaving to work on alignment — research to ensure AI systems conform to human values — at Anthropic, a start-up founded in 2021 by a group of former OpenAI researchers that bills itself as putting “safety at the frontier” of its work.
Schulman’s departure follows that of Ilya Sutskever, another of the company’s 11-strong founding group and its former chief scientist, along with other members of his team.
“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, alongside people deeply engaged with the topics I’m most interested in,” Schulman said in a note to colleagues on Monday.
“To be clear, I’m not leaving due to lack of support for alignment research at OpenAI. On the contrary, company leaders have been very committed to investing in this area,” he added.
Also on Monday, OpenAI’s president Greg Brockman announced on social media site X that he would take the rest of the year off. “I’m taking a sabbatical through end of year. First time to relax since co-founding OpenAI 9 years ago. The mission is far from complete; we still have a safe AGI to build,” he wrote, referring to artificial general intelligence, or AI that can match or better a human’s cognitive ability.
The exits, first reported by The Information, follow a tumultuous period at the San Francisco company. OpenAI has rocketed to an $86bn valuation since releasing ChatGPT in November 2022, and has rolled out a series of chatbots able to perform increasingly complex tasks.
The company has faced criticism over the pace of its AI development and direction of its research. Those concerns spilled over during a boardroom crisis last November in which Brockman and chief executive Sam Altman were ousted.
The pair returned to their roles within five days, following a campaign from employees and investors. Directors who voted to remove Altman for not being “consistently candid” have since been replaced, with Quora’s chief executive Adam D’Angelo the only director to have remained through the crisis.
However, some of the concerns raised in November have been echoed by former employees in the months since.
Elon Musk, who helped launch OpenAI but left the board in 2018 after clashing with Altman, filed a lawsuit on Monday alleging OpenAI and its chief executive had reneged on a mission to benefit humanity when they struck a commercial partnership with Microsoft, which has committed $13bn to OpenAI since 2019.
A number of other senior staff have left to pursue their own projects, with some critical of OpenAI’s approach. Jan Leike, a prominent member of Sutskever’s safety-focused “superalignment team”, left in May and criticised the company’s efforts to ensure AI could be responsibly developed and deployed.
“Over the past years, safety culture and processes have taken a back seat to shiny products,” wrote Leike, who also joined Anthropic, in a post on X.
Andrej Karpathy, another founding member of OpenAI in 2015, departed in February to work on his own AI project.
In response to Schulman’s departure, OpenAI said: “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.”
Read the full article here