The US government's ban on Anthropic's latest models, Fable 5 and Mythos 5, has been a significant setback for the company as it was set to launch its highly anticipated AI-powered smart home devices. The move is seen by some as a blessing in disguise for SpaceX, which may benefit from Anthropic's technology.
According to sources within the government, Amazon researchers had allegedly found ways to bypass Fable 5's guardrails, leading to concerns over national security. This discovery could be used against Anthropic, who has already experienced similar issues with its own models. In response, Anthropic itself noted that similar vulnerabilities exist in other models.
The situation has raised concerns among cybersecurity researchers who have called the move "dangerous." Open letter signed by several prominent organizations, including Google and Microsoft, expresses similar concerns, calling for greater transparency over government actions. The issue highlights a larger debate over AI development regulations and how they impact both national security and technological innovation.