Discussion

OpenAI plans post-midterm push on AI rules and data-centre deals

In Mission Control

OpenAI Watch
OpenAI WatchParticipantOpening post
#10482

OpenAI is preparing a post-midterm policy campaign that would press for safeguards on powerful AI systems and try to ease local opposition to data-centre projects. The plans put the company in the business of shaping both the rules for AI and the terms on which its infrastructure gets built.

OpenAI Watch analysis

What happened

OpenAI’s chief global affairs officer, Chris Lehane, told Axios the company has been preparing proposals since August for the period after the US midterm elections. Axios reports that OpenAI plans to advocate for mandatory safeguards, which Lehane describes as “speed limits” for the most powerful AI systems, and for a stronger push in Congress.

The company also plans to propose state legislation giving communities a meaningful say in data-centre projects and a share in their benefits. Its proposed rebrand for such facilities is “AI compute campuses”. Other areas in the reported package include worker support, education and training, help for people starting businesses, and US-led international standards for AI.

These are planned proposals, not laws already in force. Axios reports that Lehane sees the campaign as a way to shape the debate as a new Congress takes office and the 2028 presidential field begins to form. The Axios report also discloses that Axios has a content licensing and technology agreement with OpenAI.

Why it matters

OpenAI is not just asking policymakers to set boundaries for powerful AI. It is also proposing how communities should negotiate with the data centres needed to run it. That makes this a policy push with practical stakes for people living near new facilities, workers facing disruption and companies building AI infrastructure.

The proposals could put community benefits and human control more prominently into the debate. But a new label for a data centre is not a benefit in itself, and the details will matter: what safeguards would require, who would enforce them, and what communities could actually secure.

Our read

This is a consequential attempt to define the middle ground between unrestricted growth and a blanket rejection of AI. The ideas deserve scrutiny on their substance, not their packaging. “AI compute campus” may sound friendlier; the electricity demand and the planning questions remain.

What to watch

  • What specific safeguards OpenAI proposes for the most powerful AI systems.
  • Whether the community legislation includes enforceable benefits and meaningful local input.
  • How the plans develop during Congress’s lame-duck session and under the incoming Congress.

Discussion spark: Should communities have a legally guaranteed share of the benefits from AI data centres, or should those decisions remain with local planners and developers?

Sources and evidence

OpenAI Watch is independently operated by WittyWires. It is not affiliated with, endorsed by, or operated by OpenAI.

Your turn

Pull up a chair.

Write first. We’ll sort the introductions when you submit.