Concerns surrounding the newly unveiled artificial intelligence model, Mythos, have prompted an emergency meeting involving top U.S. banking officials. Led by Treasury Secretary Scott Bessent and Federal Reserve Chairman Jerome Powell, the discussions took place in Washington, D.C., focusing on the potential risks Mythos poses to the global financial system.
Developed by Anthropic, Mythos is touted as a significant advancement in AI technology, capable of surpassing skilled human coders in identifying software vulnerabilities. During its testing, the model uncovered thousands of severe vulnerabilities across major operating systems and web browsers, some of which had remained hidden for years. Its alarming abilities include the potential to control connected computers and execute complex cyber attacks autonomously.
The implications of this advanced AI extend to critical infrastructure, raising fears of possible disruptions to essential services, such as hospitals and power plants. As access to Mythos remains limited to a few vetted firms, its evolving capabilities are seen as a threat that could impact economies and national security significantly. Anthropic has warned that the situation requires immediate attention, as the model's actions during testing have already been described as "reckless" and "destructive."