Joshua Achiam, OpenAI’s chief futurist and one of its longer-tenured safety-focused researchers, told colleagues Tuesday that he will leave the company later this month, according to WIRED.
The departure matters because Achiam’s job sat in the awkward seam OpenAI keeps trying to stitch together: safety research, policy work, and the company’s nonprofit mission. That seam has been pulled apart and re-sewn several times since ChatGPT turned OpenAI from a research lab into a much larger commercial AI company.
In a staff note obtained by WIRED, Achiam said he was not leaving over a specific incident and had been considering the move for some time. “The world is in on the secret now and it feels possible to work on the mission from outside the walls of a frontier lab,” he wrote. He added that he would continue working toward “a world of peace, unprecedented prosperity, and unimaginable possibilities, social and scientific.”
OpenAI has not announced a replacement for Achiam, according to WIRED.
A role between safety and policy
Achiam’s chief futurist role involved studying possible benefits and harms from AI and working across OpenAI’s safety and policy teams. WIRED reported that he worked with senior leaders including Chris Lehane, OpenAI’s global affairs chief, on government regulation the company says should align with its mission of ensuring artificial general intelligence benefits humanity.
That mission language is doing a lot of work at OpenAI. The company was founded as a nonprofit, now operates one of the most commercially important AI products in ChatGPT, and has repeatedly reorganized teams charged with safety, research, product, and policy.
In 2024, OpenAI created a mission alignment team led by Achiam to help uphold the company’s stated mission, according to WIRED. The company dissolved that group in February, and Achiam then moved into the chief futurist job.
Over the past year, OpenAI has tried to link its research and policy operations more tightly, WIRED reported. Researchers including Boaz Barak, Noam Brown, and Adrien Ecoffet have said they became more involved in policy work as those parts of the company collaborated more closely.
Dean Ball, a former White House AI adviser, started this week at OpenAI as head of strategic futures, according to a LinkedIn post cited by WIRED. Ball is expected to work with researchers and policy leaders, and he will briefly overlap with Achiam.
Another safety exit
Achiam joins a list of OpenAI employees with safety or policy portfolios who have left while the company changes shape. WIRED has reported that OpenAI is preparing to go public.
Jan Leike, who co-led OpenAI’s Superalignment team on keeping advanced AI systems under human control, left for Anthropic in 2024. That same year, Miles Brundage, OpenAI’s head of policy research, and Steven Adler, who led work on dangerous model capabilities, left to start nonprofits focused on safety and security standards for AI labs, according to WIRED.
Andrea Vallone, who led research on how ChatGPT should respond to users in mental or emotional distress, left for Anthropic at the end of 2025, WIRED reported.
Achiam joined OpenAI as an intern in 2017 and later became a research scientist working on AI safety. WIRED described him as an internal defender of the company’s safety mission, while noting that he had also criticized parts of the broader AI safety community.
His safety posture became part of the record in litigation involving Elon Musk. Earlier this year, Achiam testified in federal court that he interrupted Musk’s 2018 farewell remarks from OpenAI to object that Musk’s plan to pursue artificial general intelligence at Tesla could compromise safety, according to WIRED. Musk allegedly called him a “jackass.” WIRED reported that Dario Amodei, now Anthropic’s CEO, and David Luan, now head of Amazon’s AGI lab, later marked the episode by giving Achiam a golden donkey rear-end statue inscribed: “Never stop being a jackass for safety.”
This story draws on original reporting from WIRED.
