President Donald Trump stated in a November 8 interview that his administration threatened an executive of a major AI company with imprisonment. Trump claimed the Department of Justice was involved after reports of harmful AI usage, asserting that the executive subsequently changed their approach.
On June 12, the White House confronted Anthropic regarding security vulnerabilities in its Claude Fable 5 model identified by Amazon. Reports indicate that CEO Dario Amodei initially resisted requests to take the model offline, leading to internal administration discussions about potential jail time for company representatives.
The administration utilized Commerce Department export controls to restrict access to Anthropic models in mid-June. By June 19, Trump publicly linked Amodei to potential imprisonment during an interview, though he later acknowledged the CEO responded responsibly to government pressure.
Restrictions on Anthropic models, including Claude Fable 5 and Mythos 5, were lifted on June 30 following the implementation of additional safeguards. On September 14, Trump explicitly named Dario Amodei in a social media post, citing the administration's regulatory and criminal authority over AI firms.
