US regulators warn banks of cyber risks from Anthropic's Claude Mithos model
US Treasury Secretary Bessen and Federal Reserve Chair Powell convened an emergency meeting with major bank CEOs on April 7, 2026, to warn of cyber risks from Anthropic's new Claude Mithos AI model. The model can reportedly identify vulnerabilities in operating systems and web browsers that could be exploited for cyberattacks, prompting regulators to urge banks to strengthen their defenses.
What happened
On April 7, 2026, US Treasury Secretary Bessen and Federal Reserve Chair Powell convened an emergency meeting at the Treasury Department in Washington with CEOs of major banks including Goldman Sachs CEO Solomon.
The meeting was held to discuss cybersecurity risks posed by Anthropic's new Claude Mithos AI model, which can identify vulnerabilities in operating systems and web browsers that could be exploited for cyberattacks.
Reported impact
Regulators expressed concern that AI-driven cyberattacks could become a significant risk to the financial sector. If banks were compromised, the impact could propagate through the entire financial system.
Index analysis
Regulators recognized that the Claude Mithos model's capability to identify system vulnerabilities could enable new forms of cyberattacks against the financial industry.
Governance controls
Regulators encouraged banks to strengthen their cybersecurity defenses and share their concerns about the model's potential impact on financial systems.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.