President Trump reached the point last week of viewing Anthropic as a national security threat, he said in an exclusive interview for "The Axios Show," though he signaled that relations have improved since.

Why it matters: National security concerns and personality clashes landed AI heavyweight Anthropic in the middle of a government crackdown with domestic and international repercussions.

Between the Commerce Department's imposition of sweeping export controls and the Pentagon's designating it a supply chain risk, the company has faced treatment typically reserved for foreign adversaries.

What they're saying: Axios' Marc Caputo asked Trump in a wide-ranging White House interview if he viewed Anthropic, or CEO Dario Amodei, as a threat to national security.

"Well, not now, but a week ago, maybe," the president said.
But he said he walked away from the G7 summit with the impression that Amodei was "nice" and "smart."
"He responded to us very quickly because you know it's a tremendous liability," Trump said. "People get put in prison immediately for that. You can't play games with that. And he responded very responsibly, I thought."

Catch up quick: Last week the Trump administration restricted any country outside of the U.S., and foreign nationals within the U.S., from accessing Anthropic's most advanced models.

A report from Amazon detailing a vulnerability alarmed the administration, which took it to Anthropic leadership but felt dismissed.
Technical discussions in Washington ensued. The two sides are now reportedly working on standards to evaluate AI jailbreaks.
For Anthropic, it's been a matter of learning how to communicate with the administration as much as reaching an understanding on how the technology works.

Yes, but: Trump did not rule out leveraging emergency powers under the Defense Production Act if the AI lab did not get in line, as was previously threatened during a dispute with the Pentagon.

"I have the power to use a lot of things," Trump said of the DPA. "But I'm not sure I have to do that."
"It was a competitor and a part owner that turned Anthropic in. They didn't like what they were doing. They were very concerned," Trump said of the concerns raised by Amazon. "I think so far it's been very responsible."

For the record: "We are grateful to the administration for their ongoing partnership in working to get this matter resolved as quickly as possible," Anthropic said in a statement.

"We remain committed to working alongside them towards our shared goals of protecting critical infrastructure and making sure the U.S. leads in AI."

Between the lines: The race to beat China on AI still outweighs the political clashes with Anthropic or its peers, in Trump's view.

Trump said he does not want to shut down Anthropic because the U.S. is beating China "by a lot."
"I was with President Xi. We talked about it. We're beating China by a lot," Trump said. "The good far outweighs the bad. We are going to find the bad and we're going to stop it."

The bottom line: The relationship between the White House and Anthropic appears to be on the mend.

But the technical work of setting AI safety standards and what international cooperation should look like is far less certain.