Amazon Warned on Anthropic AI Before US Crackdown

Amazon raised concerns about Anthropic's AI models before a US crackdown. Amazon CEO Andy Jassy was among tech leaders who warned senior Trump administration officials about potential security risks from Anthropic's most advanced AI models. Following a national security order, Anthropic shut down its latest models worldwide. The company had launched a public version called Fable with cybersecurity safeguards, but this was halted after government concerns about potential bypassing of these safeguards. The US government believes a "jailbreak" exists to use the model for finding cybersecurity holes. Anthropic claims this bypass only finds "minor" security flaws. In response to the order, Anthropic agreed to disable access to both Fable 5 and Mythos 5 globally for foreign nationals. A White House adviser stated the US government reluctantly issued export controls. This matters because it highlights ongoing tensions between AI development and national security.