Skip to content
NewsRegulation

Johnson Says Congress Will Not Lead on AI Safety as Trump Calls Warnings a Conspiracy

· by Pondero Newsdesk

The short version

House Speaker Mike Johnson declined to put Congress at the front of AI safety regulation on September 15, urging companies to self-regulate and calling on Trump to convene AI executives. Trump separately dismissed AI safety warnings as a conspiracy.

Johnson Says Congress Will Not Lead on AI Safety as Trump Calls Warnings a Conspiracy

House Speaker Mike Johnson stated on September 15 that Congress would not take the lead on AI safety regulation, urging AI companies to be "responsible on their own side" and calling on President Trump to convene AI executives instead of pursuing legislation. Trump, in a Fox News interview, dismissed AI safety warnings as a "conspiracy." The responses came at the end of a week in which senior researchers had left Anthropic and Google DeepMind citing safety concerns, and as tech CEOs were publicly pressing Congress to act.

What happened

Johnson addressed reporters while multiple tech CEOs were on record calling for Congressional action, per ABC News. He said that "the reflex of legislative bodies is to cover things up with red tape and hyper regulation," per ABC News. His stated position was that companies bearing safety concerns should halt their own work, and that Trump should bring AI executives together at a White House summit rather than Congress passing new requirements.

Trump's Fox News interview delivered the sharpest rebuttal to the week's safety warnings. He described the concerns as a "conspiracy" and framed China's AI development as the more urgent competitive threat, per ABC News.

CNN reported on September 13 that Trump might meet with AI executives "either this week or next week," per CNN Politics. No date for the meeting had been confirmed as of September 15.

The week's safety conversation did reach both sides of the aisle. Senate Majority Leader John Thune and Senator Amy Klobuchar had been in discussions on a bipartisan AI bill, per ABC News. House Democrats were pushing for stronger action. Both efforts face a compressed Congressional calendar before November elections and a Republican majority in both chambers that has not expressed support for new mandates.

Why it matters

No binding US federal AI safety requirements are coming in the near term. The White House, not Congress, is the potential venue for any accountability discussion, and that conversation would be voluntary.

For enterprise teams deploying frontier AI systems, the practical implication is concrete: internal evaluation gates and vendor contract terms are now the only enforceable floor. Johnson made this explicit by saying companies that believe their work poses risks should stop themselves. If your organization has been waiting for a regulatory baseline to define acceptable risk thresholds, that baseline is not arriving this year.

The week also produced a sharp reversal in direction. Multiple tech CEOs publicly asked Congress to step in. Johnson's response directed that request back at the companies themselves.

What to watch next

Whether Trump holds his potential meeting with AI executives this week, and what commitments emerge, is the next concrete signal. A Democratic legislative proposal from the House caucus discussions would need Senate agreement and a change in the administration's stated posture to move forward.

Sources