IMF Warns "AI-Driven Cyber Threats Could Shock Financial Systems"

The International Monetary Fund (IMF) has warned that advancements in artificial intelligence (AI) models could simultaneously expose and exploit security vulnerabilities in financial institutions, potentially causing shocks to the system. The IMF expressed concern that as many financial institutions rely on shared infrastructure, such as identical software and cloud services, a single vulnerability could affect multiple institutions at once.

On May 7 (local time), the IMF published an article titled "AI-Driven Cyber Attacks Are Increasing Financial Stability Risks," issuing this warning. The IMF pointed out that financial systems depend on highly interconnected, shared digital infrastructure, including software, cloud services, payment, and data networks. The IMF stated, "Advanced AI models can dramatically reduce the time and cost required to identify and exploit vulnerabilities in such infrastructure," and "they increase the likelihood of simultaneously discovering and targeting weaknesses in widely used systems." The IMF added, "As a result, cyber risks are expanding into the issue of simultaneous disruptions that can undermine financial intermediation, payments, and trust at the systemic level."

The IMF further cautioned that AI-based cyberattacks could go beyond simple security incidents and deliver shocks to the macro-financial system. If multiple financial institutions are affected at the same time, it could trigger a loss of trust, payment disruptions, liquidity pressures, and asset sell-offs. These concerns are emerging as anxiety grows over cybersecurity in the era of rapid AI development. For example, Anthropic’s next-generation AI model, Mythos, is regarded as having a powerful ability to identify software vulnerabilities, and there is market concern that this functionality could be misused.

The IMF stressed the need for international cooperation to address these risks. "Cyber risks know no borders," the IMF stated, adding, "If supervisory frameworks for AI risks are inconsistent, globally connected systems may be weakened." The IMF emphasized, "To safeguard global financial stability, it is essential to strengthen international cooperation, broaden information sharing, and enhance response capabilities."

However, the IMF also made it clear that AI could become part of the solution to cyber threats. As attackers operate at the "speed of machines," defenders must also leverage AI to detect threats and identify vulnerabilities. The IMF explained that especially by utilizing AI from the development stage to reduce security flaws, overall system risk can be lowered.

This content was produced with the assistance of AI translation services.