Heads Ministry of Treasury and Federal Reserve An “urgent” meeting was held with the chief executives of major banks to discuss concerns about the potential future cyber risks they pose AnthropicMythos and Other AI Models, Bloomberg I mentioned Thursday (April 9), citing unnamed sources.
Secretary of the Treasury Scott Besant And Chairman of the Federal Reserve Jerome Powell The report stated that the meeting was arranged at the Treasury Department headquarters in Washington to ensure that banks were aware of the potential risks and were taking necessary precautions.
Several CEOs were already in Washington for a meeting Financial Services Forumaccording to the report.
The Fed, contacted by PYMNTS, declined to comment on the report.
The Treasury Department did not immediately respond to PYMNTS’ request for comment.
According to the report, Anthropic has limited the release of Mythos to a few large companies so that they can secure their systems before releasing similar AI models.
Advertisement: Scroll to continue
It was reported on Tuesday (April 7) that Anthropic unveiled a program called Project Glasswing that will allow select partners to obtain… Early access To “Cloud Mythos Preview”.
The initiative includes participation from leading companies such as Amazon, Microsoft, and Apple, along with cybersecurity and infrastructure companies such as Crowdstrike, Palo Alto Networks, Google, and Nvidia.
The model is being developed specifically for defensive cybersecurity work, and the initiative aims to allow partners to identify vulnerabilities and harden systems before threats are exploited, according to the report.
In it advertisement Of Project Glasswing, Anthropic said: “Anthropic has also been in ongoing discussions with US government officials about the Claude Mythos Preview and its offensive and defensive cyber capabilities.”
“We hope the Glasswing project can seed a larger effort across industry and the private sector, helping all parties address the biggest questions about the impact of strong models on security,” the company also said in the announcement.
It was reported on March 26 that an accidental data leak forced Anthropic to confirm that it was testing a yet-to-be-released version. You have a model It’s called Claude Mythos and it has capabilities beyond any previously released system.
A day later, on March 27, it was reported that concerns about the capabilities of the AI model being tested by Anthropic had subsided. Cybersecurity stocks.
In November, Anthropic reported that another of its AI models had been manipulated to carry out a large-scale operation Cyber espionage practical.
For all of PYMNTS AI and digital transformation coverage, subscribe to our daily newsletter Artificial Intelligence and Digital Transformation Newsletters.
