The White House imposed sweeping export controls on Anthropic after senior officials spent roughly 24 hours trying to persuade the company to voluntarily pull a newly released AI model they believed could pose security risks.
The episode unfolded through multiple calls between Anthropic CEO Dario Amodei and senior officials in the Trump administration, including Treasury Secretary Scott Bessent, White House Cyber Director Sean Cairncross and Commerce Secretary Howard Lutnick.
The administration acted after concerns grew about the safety of Anthropic’s model and its potential to be bypassed or “jailbroken”.
Also Read | Missouri skydiving plane crashes moments after takeoff, all 12 on board killed
BREAKING: The U.S government gave Anthropic 90 minutes to shut down Fable and Mythos
“Amazon AND others” called senior administration officials to warn about models’ capabilities
Then:
1:00pm: Government calls. “Take it down.”
Cites “national security threat.” No details.… pic.twitter.com/QyYkWjsLl6— NIK (@ns123abc) June 14, 2026
The alarm bell rings in Washington
Amazon CEO Andy Jassy raised concerns to the White House on Thursday, two days after the model’s public release, after warnings that its guardrails might be bypassed.
The issue reached the highest levels of the White House by Friday morning, when Bessent, Cairncross, White House Chief of Staff Susie Wiles and other officials met to discuss the model and the government’s response.
Officials first tried to reach Amodei and were told he was unavailable, a claim Anthropic later rejected. A company spokesperson said, “This is absolutely false.”
Not quite the calamity, says Anthropic
When Amodei joined the calls, he argued that the security problem the administration was focused on was limited and specific, not a full jailbreak that would allow the model to operate without safeguards.
Amodei pushed back on the administration’s concerns and defended Anthropic’s guardrails, while Anthropic later wrote in a blog post that “no testers have yet been able to find a universal jailbreak” and that total avoidance of every jailbreak is not possible for any company.
The company also said its systems were “so strong that many users have complained that they are overly broad.”
A senior Trump administration official tells me Anthropic’s “recklessness” in responding to issues with the company’s latest artificial intelligence release led to export controls. A source close to Anthropic disputes this saying the company is eager to get this resolved and are…
— Edward Lawrence (@EdwardLawrence) June 15, 2026
The gloves remain on, yet the order arrives
The standoff ended with the administration imposing export controls on the model, which Anthropic said forced it to pull the product and “abruptly disable” it for all customers “to ensure compliance”.
A White House official told Business Insider, “Export controls were a last resort after begging them for hours to work with us. This was not something we wanted to do, but our hands were tied.”
Anthropic responded that it believed the government should be able to block unsafe deployments but said the action did not meet standards of transparency, fairness and technical grounding.
On Saturday, former White House AI czar David Sacks backed the decision, writing that the administration’s hope was for Anthropic to fix the issue so the export control could be lifted and the model returned to general release.
Also Read | ‘Deep mistrust’ remains despite deal: Iran to charge Hormuz service fees, not tolls
FAQs
Q1: Why did the White House impose export controls on Anthropic’s AI model?
Ans: The administration said it acted over concerns that the model’s safety guardrails could be bypassed, creating potential security risks.
Q2: What was the impact of the export controls on Anthropic?
Ans: Anthropic said it had to disable the affected AI model for all customers to comply with the government’s order.
