AI threatens financial stability as Mythos-style models accelerate cyberattacks, Dutch supervisors warn

Artificial intelligence (AI) is causing a structural shift in the cyber threat landscape and poses a significant risk to financial stability, Dutch regulators have warned. They point out that AI models compress attack timelines from weeks to hours and generate new vulnerabilities much faster than organizations can patch them.

The rapid development of AI technology increases the scale, speed, and complexity of cyberattacks, thus intensifying the need for better digital resilience. Because of the power and diversity of generative AI models, the likelihood and frequency of cybersecurity incidents increases.

“This raises the bar for financial institutions to address vulnerabilities swiftly, recover promptly, and implement effective fallback methods,” De Nederlandsche Bank (DNB) writes in its most recent Financial Stability Report, which was published earlier this week.

Although AI models like Mythos have proven they can find and patch unknown vulnerabilities, they can also facilitate and accelerate cyberattacks.

“For example, new (generative) AI models appear not only to be able to independently identify digital vulnerabilities in IT systems, but also to generate potential attack techniques more quickly and in a more automated manner,” the DNB’s Financial Stability Report says.

“Given these rapid developments, it remains unclear what impact such AI models will have on the nature, scale, and complexity of cyberattacks. Close and continuous monitoring is therefore essential,” the authors recommend.

According to the financial watchdog, these developments underscore the importance of a European approach, both in developing AI technology and in strengthening international information exchange about the opportunities and risks posed by these new generative AI models.

In addition, the Dutch Authority for the Financial Markets (AFM), responsible for the operation of the financial markets in the Netherlands, stated that AI is making the economy more vulnerable to cyber threats.

“Although AI applications can increase market efficiency, they simultaneously exacerbate existing vulnerabilities and introduce new risks. This, in turn, can lead to systemic risks. AI models must be reliable and secure to ensure the integrity and stability of capital markets. It is therefore important that market participants who develop and deploy AI models establish adequate safeguards, such as robust data governance and model validation,” the financial supervisor says.

The warnings from the DNB and AFM align with previous warnings from the National Cyber Security Center (NCSC).

“Response cycles measured in days are now measured in hours – weeks have become days. Organizations that fail to accelerate their patching processes, monitoring, and incident response are demonstrably at greater risk. The message to organizations: don’t treat this as just another ‘trend,’ but as a structural shift in the pace of both attacks and defense,” the cybersecurity agency warned.

Unlock more exclusive Cybernews content on YouTube.