Anthropic pulls Fable 5 over security flaws, talks with Trump admin for a fix

Anthropic, a US AI startup, had to take down its latest AI model 'Fable 5' after the Trump admin banned exports over security issues. Company reps met with US officials Monday to push for a fix. A White House official said it might take days to resolve, but didn't rule out a quick deal. Anthropic says the flaw is minor, not a big jailbreak risk. Amazon, a key investor, found and reported the bug, leading to the ban. Nearly 80 tech experts signed a letter urging the admin to reverse the ban, warning it hurts US AI leadership and market stability. Some worry this could lead to a future where every new AI model needs government approval, slowing down innovation.