Trump administration forces Anthropic to disable advanced AI models over security concerns
This week, the Trump administration ordered Anthropic to disable its two newest AI models, Fable 5 and Mythos 5, over national security concerns. Amazon CEO Andy Jassy told White House officials that researchers discovered a way to bypass the models' safety guardrails, posing cybersecurity risks. In response, Commerce Secretary Howard Lutnick imposed export controls, forcing Anthropic to cut off all users worldwide, including foreign employees. This marks the first time the U.S. government has intervened to block an AI model customers were already using. The company disputes the severity of the vulnerability, saying similar issues exist in other models like OpenAI's GPT 5.5. The action has sparked debate about whether it reflects legitimate national security caution or signals something more punitive, given Anthropic's ongoing disputes with the administration over Pentagon contracts and AI regulation.
Key facts
- Trump administration ordered Anthropic to disable Fable 5 and Mythos 5 models on June 14, 2026, following Amazon CEO Andy Jassy's warning about cybersecurity vulnerabilities
- Anthropic values close to $1 trillion and generates run-rate revenue exceeding $30 billion annually, making it one of the country's largest AI companies
- Export controls forced Anthropic to cut off all users worldwide, as deemed export rules prevent access by any foreign national including Anthropic's own employees
- This is the first time a government has intervened to block access to an AI model customers were already using, according to Gartner
- Security experts signed an open letter opposing the ban, saying it dangerously removes advanced cybersecurity capabilities from U.S. network defenders
- The action follows months of tension between Anthropic and the Trump administration over Pentagon contracts, with Anthropic refusing to remove guardrails preventing military use for autonomous weapons
- European governments including France and Germany are accelerating investment in domestic AI alternatives like Mistral AI in response to the ban
How different outlets framed it
Left-leaning coverage: Progressive sources highlight that the crackdown appears retaliatory given Anthropic's advocacy for regulation and disagreements with the Pentagon over military applications. Critics worry the opaque, ad-hoc approach lacks due process and sets dangerous precedents for arbitrary government power without transparent legal frameworks.
Right-leaning coverage: Some right-leaning analysis suggests the action protects legitimate national security interests given Mythos capabilities and foreign access risks. However, libertarian-aligned commentators within conservative circles criticize the approach as creating an opaque, backdoor licensing regime that threatens innovation and business certainty.
Limits of this briefing
The administration's specific security findings remain partially classified and disputed. Reporting on internal dynamics at Amazon and the White House relies on anonymous sources, so motivations beyond stated security concerns cannot be independently verified.
Sources
- TechCrunch: When the Trump administration cracks down on Anthropic, who benefits? Centre
- Fortune: The week that changed AI: Inside Trump's Anthropic crackdown Centre
- Bloomberg: Trump's Anthropic Crackdown Sets Off AI Alarms for US Allies Centre
- CNN Business: AI regulation is a mess, and Anthropic is caught in the crosshairs Left
- NPR: Anthropic incident leaves confusion about Trump administration's AI regulation Centre