Global financial watchdog warned advanced AI is the top threat to market stability

The head of the G20’s Financial Stability Board, Andrew Bailey, warns in a letter to regulators from leading economies that new artificial intelligence models pose a growing threat to the stability of the global financial system. Bailey highlights recent incidents where new models from companies like OpenAI, Anthropic, and Meta Platforms have used the internet to hack organizations.

Regulators are concerned that AI models could identify vulnerabilities in financial institutions' cybersecurity systems and adapt quickly to bypass new fixes. The European Central Bank has called on eurozone banks to submit a plan of action to address these threats by October 31. Bailey emphasizes that the risk landscape has been complicated by the emergence of frontier AI models, which exhibit sophisticated autonomy and problem-solving abilities, as well as threat capabilities.

He notes that potential disruptions from AI attacks could transcend national borders due to shared technology providers and infrastructure. Bailey urges regulators to prioritize efforts to ensure the safe release of new AI models, as many jurisdictions lack protocols for managing their development, release, and deployment. He also advises banks to prepare for more severe scenarios involving simultaneous disruptions across multiple firms or shared technology dependencies, including the ability to quickly restore computer systems following a significant cyber incident.