🧠 Food for thought

Implications, context, and why it matters.

A split across US government agencies shapes the NSA’s AI use

• The Pentagon gave Anthropic a formal supply-chain risk designation. Even so, the US National Security Agency (NSA), the US intelligence agency focused on signals intelligence and cybersecurity, is using Anthropic’s Mythos Preview AI tool 1.
• That exposed a divide inside the US government. Intelligence and civilian agencies want the tool for defensive cybersecurity 2.
• In an April 14 email to Cabinet officials, the White House Office of Management and Budget (OMB), which helps oversee federal agency policy and spending, said it is setting up safeguards so major federal agencies can access Mythos. It did not give a timeline 2.

Mythos is raising global concern about cyber risk to the financial system

• Mythos can find and exploit software bugs on its own. Governments and regulators are treating that as a possible threat to the global economy 2.
• At International Monetary Fund meetings, finance ministers and central bankers discussed the model as a serious risk to the resilience of global financial systems 2.
• The concern has moved past theory. The US Treasury has urged major banks to test their systems against Mythos 2.
• The Barclays CEO and the Bank of England governor have both spoken in public about assessing the risks and vulnerabilities tied to the model 2.