Innovation & Tech

Jacob Coxon to sell OpenAI equity and donate proceeds to AI safety

Former OpenAI researcher Jacob Coxon plans to sell his equity, surrendering a profit that could reach millions, and donate the proceeds to AI safety after warning about the superintelligence race.

By Nathan Brooks

3 min read

Former OpenAI researcher says he’ll donate his equity to AI safety: 'Everything else is a distraction'
Former OpenAI researcher says he’ll donate his equity to AI safety: 'Everything else is a distraction'WOKANDAPIX / pixabay

What's News

  • Coxon committed Wednesday to sell his OpenAI equity and donate the proceeds to AI safety causes, potentially giving up a profit worth millions of dollars.
  • Coxon, 27, worked at OpenAI from 2023 to 2026 before joining and recently quitting Anthropic.
  • His Sept. 8 X post, which said AI could kill “us all” by the end of the decade, has received 175 million views.
  • At a Nov. 9 Stanford talk, he called his OpenAI stake “blood money” but resisted giving it up before changing course.
  • Coxon testified at a New York City Council AI-safety hearing on Oct. 5, 2026.

By Nathan Brooks

Jacob Coxon said on Wednesday that he was committed to selling his OpenAI equity and donating the proceeds to AI safety causes, giving up what could be millions of dollars in profit. The pledge comes as OpenAI CEO Sam Altman has said an initial public offering is on the horizon.

Business Insider reported that Coxon, 27, is from England and worked at OpenAI from 2023 until 2026. He then joined Anthropic and recently quit. Since then, he has made the media rounds to advocate for AI safeguards and, principally, a pause on recursive self-improvement, the process by which AI models train themselves.

Coxon has pressed the case in repeated public statements. “We need to pause or drastically slow down the race to superintelligence,” he said on X. “Everything else is a distraction.”

Business Insider reported that Coxon’s warnings sparked a national conversation in September, when he left Anthropic and warned that it and OpenAI were “gambling with our lives.” On September 8, he wrote on X: “The people building AI earnestly believe that it could kill us all by the end of the decade.”

The post has since received 175 million views. Coxon added: “This is not a marketing stunt.”

The financial trade-off is substantial. Business Insider said Coxon could otherwise have made millions of dollars in profit when his OpenAI equity became liquid at a public listing, though the report gave no exact valuation, offering date or IPO structure.

His position was different in a November 9 talk at Stanford. Coxon called his stake in OpenAI “blood money” but balked at giving it up at the time.

“I like money, bro,” he said. “My parents were like worried I was having a psychotic break. I think if I'd also given away all the money they'd be even more convinced of that.”

That reluctance has now ended, at least in stated intent. Coxon’s Wednesday commitment earmarks the equity’s sale proceeds for safety work.

Business Insider described an individual sale, not a company-level donation. It depends on Coxon selling his own private holding, and the report did not specify a buyer, sale route, recipient or guaranteed proceeds. For OpenAI employees and investors, the move makes the tension between private incentives and public safety claims visible. The pledge concerns one researcher’s holding, not an OpenAI contribution or employee-share policy.

Coxon’s position also sits within a broader regulatory contest. A Business Insider photograph showed him testifying at a New York City Council Committee of the Whole hearing at City Hall on October 5, 2026. Its caption said top executives from OpenAI, Anthropic, Google and Meta were testifying under oath about AI safety risks and proposed municipal regulations.

What does this change for AI safety teams?

It creates a potential new pool of capital for AI safety research. Because Business Insider did not provide an exact valuation, sale schedule, recipient or guaranteed amount, safety teams cannot yet treat the pledge as a sized funding commitment.

For practitioners, the immediate change is a possible increase in private resources devoted to safety work. The move also gives donors and regulators a concrete test: Coxon has aligned his own financial outcome with his demand for slower development rather than leaving that conflict rhetorical.

That alignment does not settle the technical debate over recursive self-improvement. The source provides no evidence of a change to OpenAI’s strategy because it describes an individual’s equity sale, not a corporate contribution, employee-share restriction or change to the IPO plan.

His reversal will be tested against his earlier warnings. The former researcher called his OpenAI holding “blood money” and openly said he liked money, but now says the proceeds should go to safety causes.

The next measure is disclosure: valuation, buyer or sale route, recipient, and amount. Until those details emerge, the pledge adds visibility to AI safety without changing the terms of the development race Coxon wants slowed.

Source: feeds.businessinsider.com

Share this article:

More from Nathan Brooks

Nathan Brooks

Show full bio

News editor covering regulation, policy and their impact on entrepreneurship, startups and venture funding at Business Bearings.

2,426 articles

Related articles

« Previous articleNext article »