In a dramatic escalation of government oversight over artificial intelligence, the U.S. Commerce Department sent Anthropic an enforcement letter late Friday that effectively forced the company to pull its two most advanced AI models, Fable 5 and Mythos 5, offline. The letter invoked an obscure export control directive, banning non-Americans—including Anthropic's own employees—from accessing the models, citing an unspecified national security concern. Anthropic said it believes the letter is related to a bypass of the model's guardrails, but the company has not received specific details, and the letter remains under seal.
Anthropic promptly shut down both models to all customers to ensure compliance, marking the first time the U.S. government has used such a directive to disable a major AI product. The swift and unilateral action did not appear to require court approval, raising alarms about executive overreach. The Trump administration's intervention demonstrates that even the most cutting-edge AI labs are not immune to government interference, and it serves as a stark warning to the entire tech industry: comply, or risk being shut down.
A Misunderstood Security Report
New details that emerged over the weekend cast serious doubt on the government's already shaky reasoning. Cybersecurity veteran and researcher Katie Moussouris, founder of Luta Security, revealed in a blog post that Anthropic had recently shared with her a private copy of a paper written by security researchers at Amazon describing an alleged guardrail bypass in Fable 5. The paper detailed how the researchers triggered a bypass by asking the model to "review code for security issues" versus asking it to "fix this code." The end result is largely the same, even if the questions are phrased differently.
Moussouris argued that the bypass "should never have triggered an export control." She explained that the behavior described in the paper cannot meaningfully be fixed, and any attempt would only weaken the model for defense. She criticized the export control directive as hasty, heavy-handed, and misguided. Moussouris and dozens of other top security researchers and experts have since called on the Trump administration to revoke the order, calling the move to pull advanced cybersecurity capabilities from network defenders in the U.S. as "dangerous."
The Real Reason Behind the Ban
Axios, citing sources, described a tense situation over the weekend between Anthropic and the administration. Personality differences between Anthropic and the Trump administration appear to have driven the export directive, rather than a genuine technical threat. The existing fractious relationship between the two parties made the government more willing to act without thorough investigation.
Past administrations have made sweeping decisions on knowledge gaps. For instance, language used by the U.S. government during the 2010s to fix export law covering cybersecurity tools that could also be used for cyberattacks was so broad that it inadvertently nearly outlawed legitimate security and vulnerability research. The Trump administration's directive appears retaliatory, lacking transparency and proportional justification.
Justin Hendrix, editor of Tech Policy Press, said the move is "likely to raise alarms in foreign capitals about the reliability of American AI for critical applications." The message sent to the world is that AI companies in the United States cannot be trusted to operate without political interference from the U.S. government.
Unanswered Questions and Dangerous Precedents
The Trump administration has not confirmed why it invoked the export control directive. Did officials misread the security report and overreact? Did Amazon CEO Andy Jassy—whose competitors might benefit from Anthropic's downfall—say something to senior government officials that prompted the reaction, out of caution or spite? Was something lost in translation, or was this a way to pressure a company with which the administration already has a fractious relationship? It is entirely possible that the White House was unaware of the far-reaching consequences of the letter’s demand and officials are now scrambling to undo the damage of their own making.
The aftermath sets a dangerous precedent about how much control the government intends to wield over the release of American-made software. This time the government took issue with Anthropic; tomorrow it could be with anyone else. The climate, as Hendrix puts it, is one of a cloud of suspicion that senior officials are picking favorites based on personal and political factors. The AI industry is now on notice: compliance is mandatory, and the government’s patience is thin.
The implications are profound. For network defenders, losing access to advanced AI tools like Fable 5 and Mythos 5 means a reduction in cybersecurity capabilities. For the broader tech ecosystem, the uncertainty created by such ad hoc government interventions could chill innovation and push companies to relocate operations overseas. The United States, long the leader in AI development, risks ceding ground to countries that offer more predictable regulatory environments.
The security community has rallied behind Anthropic, but the damage is done. Unless the administration revokes the order and provides clear justification, the trust between the tech industry and the government will continue to erode. The first StrictlyVC of 2026 hits SF on April 30, but for now, the spotlight remains on the fragile relationship between Silicon Valley and Washington.
Source: TechCrunch News