ECB to ask banks for targeted measures to counter AI risk

FRANKFURT, June 3 (Reuters) – The European Central Bank met commercial lenders last week to discuss the potential threat from the newest artificial intelligence models and will follow up with demands for practical defence measures, board member Frank Elderson said.

Rapidly advances in AI technology have increased the risk that cyberattacks can rapidly discover and exploit vulnerabilities in bank defences and combine seemingly minor issues into serious threats.

“As a next step we will send a so-called ‘dear CEO letter’ to all banks in which we aim to ask banks to take proactive measures to ensure the continued robustness and security of their systems in the face of these transformative challenges and will follow up with individual banks in a targeted manner,” Elderson said on Wednesday.

Elderson, also the Vice Chair of the ECB’s Supervisory Board, argued that such threats are not merely a cybersecurity issue and require a strategic approach with management taking charge, devoting expertise and resources over the span of many years.

This may be expensive but high profitability in the sector should give lenders the financial buffers to invest, he argued.

A key issue may be that large banks can more easily afford defences, this may be a stretch for small and medium sized lender.

“The critical infrastructure on which banks depend – including cloud providers, telecommunications networks, payment systems and electricity and water supplies – could also become targets,” he said. “As a result, scenarios that were once considered tail risks may become more likely.”

(Reporting by Balazs Koranyi; Editing by Toby Chopra)