US Treasury Secretary Scott Bessant and Federal Reserve Chairman Jerome Powell called an emergency meeting with bank executives this week to warn them about the cybersecurity risks posed by a new artificial intelligence model from Anthropic.
US Treasury Secretary Scott Bessant and Federal Reserve Chairman Jerome Powell called an emergency meeting with bank executives this week to warn them about the cybersecurity risks posed by a new artificial intelligence model from Anthropic.
This was reported to Reuters by two sources familiar with the situation.
Anthropic launched its powerful Mythos model in a limited release earlier this week amid concerns that it could expose previously unknown vulnerabilities in cybersecurity systems. The developers believe the model is capable of detecting and exploiting vulnerabilities in «all major operating systems and all major web browsers».
Last week, Anthropic said it was in ongoing talks with US government officials about the model’s «offensive and defensive cyber capabilities».
The Treasury Department held meetings in Washington to make sure banks were aware of the risks associated with Mythos and similar models and were taking steps to protect their systems, one of the sources said. The invitations came while most of the heads of the largest U.S. banks were already in Washington for other meetings.
About 40 technology companies, including Microsoft and Google, currently have access to Mythos. Anthropic has previously reported concerns about the speed of cyberattacks that can be carried out using Mythos.
