Amazon CEO Raised Concerns About Anthropic Models Before US Government Crackdown

Following concerns raised by Amazon CEO Andy Jassy, the Trump administration has directed Anthropic to block all foreign access to its advanced AI models, Mythos 5 and Fable 5, citing national security risks. Anthropic complied by disabling the models globally. The move, which follows public safety warnings from Anthropic's own CEO, has drawn criticism from some AI researchers as an overreaction.

Mythos model security concerns

• ▪Jassy told officials that Amazon researchers prompted Anthropic's Fable 5 model to provide information that could aid cyberattacks
• ▪The U.S. government believes a "jailbreaking" method exists that would allow Fable 5 to be used to identify software vulnerabilities
• ▪Andy Jassy's conversations with U.S. officials, including Treasury Secretary Scott Bessent, prompted the crackdown on Anthropic's models
• ▪Amazon CEO Andy Jassy raised concerns to senior Trump administration officials about security risks in Anthropic's most advanced AI models

US government access restrictions

• ▪The U.S. government's restrictions on Anthropic's models were implemented in the form of an export control
• ▪The Pentagon's chief information officer expressed support for the government's action in a post on X
• ▪In response to the government order, Anthropic disabled global access to its Mythos 5 and Fable 5 models
• ▪On June 13, 2026, the Trump administration directed Anthropic to block foreign access to its Mythos 5 and Fable 5 models, citing national security
• ▪The order blocks all foreign nationals, whether inside or outside the U.S., from using the models

Industry criticism of crackdown

• ▪AI pioneer Yann LeCunn attributed the crackdown to Dario Amodei's "ridiculous fear mongering about Mythos/Fable."
• ▪AI researcher Gary Marcus called the administration's move "wildly overdramatic and also counterproductive."
• ▪Some experts criticized the action for affecting allied nations, with Jimmy Goodrich of the University of California calling it "not well thought-out."

Amodei's safety positioning

• ▪Dario Amodei left OpenAI to found Anthropic over concerns that OpenAI prioritized releasing products over ensuring their safety
• ▪Anthropic CEO Dario Amodei had publicly warned that his company's Mythos model presented "very real risks" to cybersecurity and national security
• ▪In a June 2026 essay, Amodei warned that biological and AI autonomy risks could follow the current cyber risks from advanced models

Anthropic's pause advocacy

• ▪Anthropic recently called for a temporary halt to frontier AI development to allow safety research and societal structures to catch up
• ▪An Anthropic paper cautioned that new models are getting closer to self-improvement, which could increase risks of humans losing control