Trump on AI: whoever wins AI, wins

As top AI lab CEOs call for a slowdown, Trump and House leaders say the real risk is losing to China.

Sep 13, 2026

6 cards · 1m 40s · tap to begin

From · · 1m 40s

Trump on AI: whoever wins AI, wins

As top AI lab CEOs call for a slowdown, Trump and House leaders say the real risk is losing to China.

In brief

As top AI lab CEOs call for a slowdown, Trump and House leaders say the real risk is losing to China. Washington's frame is the race, not the risk, so any real slowdown will have to start inside the labs themselves. Originally reported by NPR.

Trump waves off the warnings

U.S. President Donald Trump attends the Irish Open golf tournament at Trump International Golf Links in Doonbeg, Ireland, Sunday, Sept. 13, 2026.

Speaking to reporters at his Doonbeg golf resort in Ireland on Sunday, Trump called the dire predictions about AI going rogue exaggerated, the work of what he called "negative forces." He never said who he meant.

“You have a lot of negative forces that are bringing it up that shouldn't be bringing it up.”

The race is the point

His argument rests on one line: America leads China in AI and he intends to keep it that way. Guardrails are fine with him, as long as they sit inside that goal rather than slow it down.

“We're leading China in AI, we're the most sophisticated country in the world... whoever wins AI, wins.”

No moratorium, says Johnson

House Speaker Mike Johnson told CNN's State of the Union that an emergency session to regulate AI would hand the race to China. He wants tech companies, not Congress, writing the safety rules.

“If Congress just races in and does some sort of emergency session to try to regulate AI, we will lose the race to China.”

Amodei asks for brakes

Anthropic CEO Dario Amodei posted an essay Saturday urging companies and governments to slow how fast AI capabilities improve, warning about loss of control plus cyberattack and bioterrorism uses.

Musk and Altman sign on

Both said they share Amodei's concerns. Altman told Fortune he would delay OpenAI's IPO until at least 2027 over safety, and that the company has discussed pausing tests to make the technology safer.

A White House adviser shrugs

David Sacks of the President's science advisory council told the CEOs to stop asking permission and just slow down themselves, while questioning how much of the caution is really about liability.

“You face massive product-liability exposure if your products enable a truly damaging cyberattack.”

The key point

Washington's frame is the race, not the risk, so any real slowdown will have to start inside the labs themselves.

Read the original on NPR

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 policy and related topics.