Italy's Largest Private Bank Loses Over $100M to AI-Assisted Fraud

Fideuram, Italy's largest private bank and part of the Intesa Sanpaolo group, saw about 95 million euros (more than 100 million US dollars) transferred out after fraudsters used AI-generated messages and a cloned voice to impersonate people its chairman trusted.

How the fraud worked

In February 2026, Fideuram's then chairman, Paolo Molesini, received a WhatsApp message that appeared to come from Carlo Messina, CEO of parent company Intesa Sanpaolo. He then received a call from someone posing as a corporate lawyer he knew. The lawyer's voice had been cloned with AI tools.

Believing the instructions were genuine, the chairman had the CFO authorize a series of transfers. Most of the money went to accounts in China and Hong Kong.

Cooperation between authorities in Italy, China and Portugal recovered about 53 million euros. The remaining 36 million euros was moved through overseas accounts and converted into cryptocurrency. Milan prosecutors are investigating a foreign national living outside Europe.

Why it matters

This is one of the clearest cases showing that AI voice fraud has moved from theoretical risk to real financial loss at scale. The victim was not an inexperienced individual but the top executive of a bank with strict controls.

What companies should do

The simplest rule is also the most effective: large approvals should never rest on a single call or message.

• Verify every urgent payment request through an independent second channel, for example by calling back a number already on file, not one given in the message.

• Require two-person approval above a set threshold.

• Train finance, accounting and executive assistant staff to spot voice and video impersonation.

• Agree on an internal verification phrase between executives and the finance team.

Source: VnExpress, Reuters