OpenAI, Anthropic researchers now steer company policy

An administration source told Axios that OpenAI's own researchers had "neutered" the company's Washington influence, leaving it "no longer seen as or acting as a leader in driving policy."

Per Axios, elite scientists at OpenAI and Anthropic are now "wielding unusual influence over corporate policy," pushing executives toward safety stances and government oversight. The leverage is money: compensation packages have climbed "into the hundreds of millions of dollars," with favorable vesting and freedom to discuss their work publicly.

The clearest case is OpenAI President Greg Brockman's abandoned donation to Leading the Future, a pro-AI super PAC. Brockman and his wife had pledged $50 million; the first $25 million was paid in 2025, and the second $25 million will not be sent. In a Slack note on June 11, Brockman wrote: "I feel terrible about the fact that LTF keeps reflecting on the company and become so much of a distraction for people here." By mid-July, eight OpenAI employees had collectively donated more than $215,000 to Guardrails Alliance, a rival PAC.

The pressure extended to statehouses. Researchers reportedly moved OpenAI to back a transparency measure in Illinois and an auditing bill in California late in those legislative sessions. More than 1,000 workers from OpenAI, Anthropic and Google signed the "Pacing the Frontier" petition, which asked the U.S. government to help "deliberately pace the frontier of automated AI development."

A White House official pushed back on the framing, calling OpenAI "an incredible partner" and pointing to a recent accord the company signed with the president. At Anthropic, researchers reportedly complicated the company's efforts to reach an agreement with the Pentagon after a confrontation with senior administration officials. The pattern lands amid a run of senior safety departures we have been tracking at OpenAI.