Washington won't regulate AI anytime soon

Trump killed a proposed AI oversight agency, Congress is gridlocked, and tech executives are blocking most proposals.

By ·

Sep 16, 2026

7 cards · 1m 50s · tap to begin

From · · · 1m 50s

Washington won't regulate AI anytime soon

Trump killed a proposed AI oversight agency, Congress is gridlocked, and tech executives are blocking most proposals.

In brief

Trump killed a proposed AI oversight agency, Congress is gridlocked, and tech executives are blocking most proposals. Despite mounting concerns about frontier AI risks, the US government has no serious regulatory plans in motion and faces political obstacles that make action unlikely in the near term. Originally…

Trump killed the AI oversight plan

A turtle with the US Capitol dome as a shell and a pixelated rabbit jumping over it

The White House was developing a FINRA-style regulatory body for AI labs, but Trump soured on the idea after tech executives including David Sacks and Mark Zuckerberg called to oppose it in August.

Trump calls AI safety a hoax

Trump has rejected any overt regulation and called AI safety concerns a hoax on Truth Social, partly in response to Anthropic CEO Dario Amodei's essay on frontier AI dangers.

Congress has competing bills with no votes

Several AI bills exist in Congress, but with no voting days left before midterm elections and no unified support, passage is extraordinarily unlikely.

The Frontier Act has the most support

The Frontier Act has the most support

The bipartisan Frontier Act, introduced by House Democrat Lori Trahan and Republican Jay Obernolte, would embed auditors at AI labs to report safety incidents. OpenAI backs it, but Speaker Mike Johnson won't bring it to the floor.

China concerns block the auditor idea

Critics worry that if auditors force a model offline, China won't pause development, creating pressure to lift emergency pauses to stay competitive. This defeats the purpose of having auditors.

Kill-switch bills lack tech support

Two competing kill-switch proposals would let the government shut down AI models, but tech executives are skeptical. They argue kill switches can't stop damage already done by unobserved models.

No serious plans on the horizon

Despite widespread talk of AI regulation, there are no new oversight proposals seriously under consideration at the White House, and Congress faces insurmountable obstacles to passing anything.

The main idea

Despite mounting concerns about frontier AI risks, the US government has no serious regulatory plans in motion and faces political obstacles that make action unlikely in the near term.

Read the original on WIRED

React

Sign in to react and comment.

Comments (0)

Life is short. Keep it sweet. Respect others' opinions and be kind!

    Recommended next

    More decks on ai regulation and related topics.