OpenAI's 'compute campus' push faces scepticism as lawmakers demand more control over AI

OpenAI is trying to reset its policy pitch after the midterm elections, but its effort to soften the debate around data centres and artificial intelligence is facing resistance from lawmakers and safety advocates.

The company is promoting the term “compute campuses” for data centres and says communities should have a voice in how those facilities are built. It is also comparing AI regulation with speed limits for cars, arguing that AI already supports people in daily life while calling for global standards.

The strategy, first reported by Axios, is intended to present OpenAI as a practical alternative to both unrestricted development and tougher opposition to the industry. Early reactions suggest the company still has substantial work to do to persuade voters and Democrats that its policy positions reflect more than an attempt to improve its public .

Rebranding meets resistance

Brad Carson, a former Oklahoma Democratic congressman and co-founder and president of Americans for Responsible Innovation, said the language change would not persuade critics.

“If they think they can change the AI debate with Orwellian name changes, they are setting themselves up for failure,” Carson said. His organisation supports stronger AI guardrails.

Rep. Greg Casar, D-Tex., also criticised OpenAI's political approach in a on X. “AI will be the number one issue in 2028,” he wrote. “And anyone looking to OpenAI's lobbyist for advice will lose.”

OpenAI's top lobbyist, Chris Lehane, nevertheless argued that the company's updated positions could prove durable, even if they leave it in what he described as a “lonely position”.

Lehane told Semafor that hundreds of state AI bills could emerge in November, December and January. He said OpenAI plans to support a mandatory AI testing bill capable of attracting 60 votes by bringing lawmakers from both parties together.

OpenAI's proposed policy path

OpenAI continues to support some state-level AI legislation. Lehane said the company recently endorsed two frontier AI safety bills in California.

He also said OpenAI had discussed Rep. Ted Lieu's AI “kill switch” bill, which the California Democrat proposed in July. Lehane said the company would soon have “next steps” on measures to make its products safer for teen users.

Lehane presented OpenAI's approach as a middle path. He argued that outright opposition to data centres could leave the United States at a disadvantage to China, echoing the White House's argument for a light-touch approach to AI regulation.

That argument has not convinced all critics. Alex Bores, a New York State assemblyman who became a leading target of the AI industry during his run for Congress, said Lehane was seeking the best possible arrangement with the Trump administration before Democrats gained more power.

Bores also acknowledged one point in OpenAI's message: the debate is ultimately about control, and the public needs a stake in decisions about the technology.

Why the next phase may require compromise

Any effort by Democrats to advance AI policy next year would require cooperation with Republicans, regardless of which party controls Congress. That makes compromise and engagement with AI companies likely to remain central to the legislative process.

Joseph Hoefer, chief AI officer at Monument Advocacy, told Semafor that other technology companies could follow OpenAI's lead. He said political calculations were moving in that direction because polling showed limits to the industry's earlier lighter-touch strategy.

A hands-off message becomes harder to maintain when voters are asking who controls AI, Hoefer said. For companies seeking continued growth in the sector, he argued, the challenge is to give lawmakers an argument for AI investment that they can defend in their home communities.

For readers, the dispute points to a broader policy question: whether AI companies can win public support by changing their language and supporting selected safety measures, or whether lawmakers will demand stronger oversight before approving further expansion.

Conclusion

OpenAI is combining community engagement, selected safety legislation and a moderate regulatory message in an effort to influence the next phase of AI policy. The company's lobbyists face scepticism from lawmakers and advocates who want clearer public control over the technology.

Frequently Asked Questions

Q. What is OpenAI calling its data centres?

OpenAI is calling them “compute campuses”. The company is also seeking community input on their construction.

Q. What broader policy message is OpenAI promoting?

OpenAI is comparing AI rules with speed limits for cars and supporting global AI standards. It is presenting its approach as a middle path between unrestricted development and opposition to data centres.

Q. What AI legislation does OpenAI plan to support?

Chris Lehane said OpenAI plans to push for a mandatory AI testing bill that can attract 60 votes by building bipartisan support.

Q. Has OpenAI supported state-level AI legislation?

Yes. Lehane said OpenAI recently endorsed two frontier AI safety bills in California.

Q. What did OpenAI discuss with Rep. Ted Lieu?

Lehane said OpenAI spoke with Lieu about the AI “kill switch” bill he proposed in July. He also said the company would soon have next steps on improving product safety for teen users.