Anthropic's AI Ban: More Than Just Amazon's Warning, Report Suggests

The sudden ban on Anthropic’s Fable and Mythos AI models by Trump administration may not have been triggered solely by Amazon’s warning, according to a report by Axios. Analysts and officials are of the opinion that the company’s inability to effectively communicate with the White House and perceived disregard for a recent cyber executive oder played a major role. “Everybody said Anthropic was a bad actor… They screwed us,” one administration official said. On June 11, Amazon CEO Andy Jassy raised alarms with the Treasury Secretary Scott Bessent, warning that Anthropic’s frontier models could be jailbroken. Administration officials claim that Anthropic knew that such vulnerabilities existed but still chose to distribute the models. By the night of June 12, the government imposed stringent export controls, forcing Anthropic to take the system offline for all non-US users.

Ideological and personality clashes

As per the Axios report, behind the scenes, sources say that Anthropic has struggled to “speak the administration’s language.” One insider described the relationship as “like they just speak in different languages,” noting that personality differences and optics worsened tensions. Also, the company’s dismissive blog post about Amazon’s complaint, coupled with its decision to enlist a cybersecurity expert viewed by the administration as politically hostile, further fueled mistrust.

National security concerns

The White House had already threatened export controls weeks earlier after learning Mythos was accessed by a foreign entity linked to the Chinese Communist Party. Although Anthropic revoked access in that case, officials felt the company’s response was insufficient. The Pentagon had also clashed with Anthropic in the past, with negotiations breaking down over personal disagreements.

What’s next for Anthropic

The Commerce Department is scheduled to meet with Anthropic’s senior tech staff, including Logan Graham, Dave Orr, and Nicholas Carlini, to discuss compliance with the cyber executive order. Additional meetings with the CIA and White House science advisor Michael Kratsios are planned. Officials say one option is to ensure Anthropic’s models cannot be jailbroken — though perfect resistance may be impossible. Another is to fix what one source called an “attitude problem,” so regulators feel secure.

Anthropic's Mythos may have suffered an 'embarrassing' leak

In related news, A China-linked group may have gained access to Anthropic's powerful Mythos AI model, according to a report by Semafor. The reported access is said to be one of the factors that may have influenced the White House's decision to impose export restrictions on Anthropic's latest AI systems, Mythos 5 and Fable 5. While the US government has not officially confirmed the claim, the report has raised fresh questions about the security of advanced AI models and whether potential leaks played a role in the restrictions that effectively cut off foreign access to the technology. According to the report, the White House was concerned that a group linked to China may have accessed Mythos. It said such access could pose a national security risk because advanced AI models can be used for cybersecurity research and vulnerability discovery. Officials were also reportedly concerned that access to the model could allow attempts to replicate its capabilities through AI distillation, a process in which another model learns from a more advanced system.