That warning came from European Central Bank Governing Council member José Luis Escrivá, who spoke at an event in the Spanish city of Tarragona on Saturday (May 9).
“Recent developments in artificial intelligence force us to reassess the robustness of our financial infrastructure and our cybersecurity,” said Escrivá, whose comments were reported by Bloomberg News.
As that report noted, Anthropic’s Mythos model has triggered fears of a new era of cyberattacks. The AI platform is a source of concern for regulators, central bankers and corporate leaders, who are seeking more insight into the new tool which has yet to be widely released.
In addition, there’s a concern that financial systems in Europe and other places outside the U.S. are disadvantaged by limited access to Mythos.
“We are living through an unprecedented wave of technological change that’s redefining the very fabric of our economies, but also bringing new risks,” Escrivá said, including the potential impact of private stablecoins.
Advertisement: Scroll to Continue
“Private stablecoins, by their nature, cannot anchor the monetary system: Their stability rests on contingent confidence in their backing, and that confidence can vanish at the very moment it’s most needed,” said Escrivá, who is also a governor for Spain’s central bank.
“What central banks provide — and what no private instrument can substitute — is the ultimate anchor of trust on which financial systems rest.”
Escrivá’s warning on AI came two days after the U.S. Office of the Comptroller of the Currency (OCC) cited AI as a possible threat in its spring 2026 Semiannual Risk Perspective.
“Artificial intelligence is significantly transforming the cyberthreat landscape, while also providing new capabilities to manage cyber-related risks,” the regulator said.
The OCC called on banks to mitigate AI-enabled cyber risks by creating more stringent security measures, such as multifactor authentication and timely patch management. The report also recommended that banks deploy AI to counter threats, and said lenders should understand the advantages and risks of the increasingly advanced AI tools coming to market.
“Artificial intelligence has opened up Pandora’s box for enterprise cybersecurity,” PYMNTS wrote recently. “And what it found was that the modern enterprise is no longer a closed system. It is a web of dependencies, stitched together by software vendors, cloud providers, and outsourced engineering partners. Increasingly, this means the weakest link isn’t one that’s found inside the organization at all but instead resides across the long tail of third-party software that keeps operations running.”
