Global Financial Regulators Assess Risks of Anthropic Mythos AI to Banking Systems

Key facts

• The Australian Securities and Investments Commission (ASIC) and the Australian Prudential Regulation Authority (APRA) are monitoring Anthropic’s Mythos AI model for its potential to destabilise banking systems.
• South Korea’s Financial Supervisory Service (FSS) and the Financial Services Commission (FSC) convened with banks and insurance companies to review vulnerabilities associated with the AI tool.
• The National Cyber Security Center (NCSC) in the Netherlands reported that Mythos can construct chain attacks by combining small, seemingly harmless software bugs into serious exploits.
• The Canadian Financial Sector Resiliency Group (CFRG) and the Office of the Superintendent of Financial Institutions (OSFI) met with executives from Canada’s six largest banks to discuss AI-related cybersecurity risks.
• Anthropic has restricted Mythos from public release to prevent the tool from being weaponised to detect and exploit software vulnerabilities within the global financial sector.