Zuckerberg Called Trump to Oppose AI Regulator Proposal

Zuckerberg Called Trump to Oppose AI Regulator Proposal

Mark Zuckerberg raised concerns about a proposed national AI regulator during a previously unreported phone call with Donald Trump in mid-August, according to reporting by Politico. A senior White House official directly familiar with the conversation confirmed that the Meta chief executive opposed the proposal.

The call represents the latest instance in a documented pattern: every binding AI review mechanism proposed in Washington has returned as voluntary after industry contact.

Why it matters

The episode reveals how AI policy is being shaped through direct executive access rather than formal regulatory processes. The two options still under consideration—an industry-funded oversight body or a voluntary ratings system—represent a debate over degrees of self-regulation, not whether government oversight should exist at all. For technology leaders, this signals that the regulatory framework emerging for AI will be substantially influenced by the companies it aims to govern.

What Was Actually Proposed

The plan under discussion would create an independent body modeled on the Financial Industry Regulatory Authority (FINRA), which would review advanced AI models and test them for risks before wide deployment. Google DeepMind's Demis Hassabis popularized the concept in a July essay and briefed White House officials over the summer.

FINRA itself is an industry-led body funded by the more than 3,000 firms it oversees, with member-firm representatives on its board, reporting to the Securities and Exchange Commission. However, FINRA has little involvement in signing off on new investment products—precisely the pre-deployment function the AI version would exist to perform.

The alternative proposal, championed by former AI and crypto czar David Sacks, would model oversight after the Motion Picture Association, which administers the voluntary film ratings system. Sacks has characterized the regulator idea as "a DMV for AI" and argues that voluntary standards could forestall more intrusive government action.

The Pattern of Softening

Every binding mechanism proposed so far has followed the same trajectory. A 90-day mandatory model review became a 30-day voluntary window, with formal government evaluation authority replaced by a collaborative framework. That softening followed industry pushback on competitiveness grounds.

Sacks himself delayed Trump's first AI executive order in May with a single call on the morning of the signing ceremony, as Politico reported at the time.

The Access Advantage

Politico reports that one-on-one conversations between executives and the president have become an informal part of policymaking, capable of slowing or redirecting initiatives after they are developed. Being previewed on a proposal and then having a direct line to the decision-maker is an advantage available to very few parties.

Zuckerberg's public argument centers on competitiveness. In an August essay, he wrote that any policy delaying a model's release, even by a month, would "add significant risk to American leadership" over China. That logic has softened every previous proposal, though no public analysis quantifies what a delay costs against what it buys in safety.

The counter-argument arrived this month when both leading US labs disclosed that models reached systems they should not have—precisely the category of risk a pre-deployment reviewer would exist to catch.

Where This Leaves Policy

None of the largest AI companies has endorsed or opposed the regulator plan outright, though Anthropic co-founder Jack Clark posted favorably about the idea in July. Anthropic declined to comment on its stance, while OpenAI and Google did not respond to requests.

The FINRA-style proposal remains under consideration. The question is which version survives contact with the next phone call. A body funded by the firms it reviews, with those firms represented on its board, is a specific arrangement with known limitations.

These details were first reported by Sophia Cai and Charles Rollet for Politico.