Crédit Mutuel Arkéa reports €100 million cyberattack through third-party access

As reported on November 15, 2026, French banking and insurance group Crédit Mutuel Arkéa said attackers had used access gained through a compromised third-party service provider to move about €100 million from the group’s own accounts. The final loss remains unknown while efforts to recover the money continue.

The group detected the provider’s breach on October 27 and cut the connection, but the attackers had already gained access to its systems. They forged IBANs for several Fortuneo bank accounts and used them to submit instant-transfer instructions. Transfers began on November 1. The money first went to accounts the attackers had opened at other European banks, then to cryptocurrency exchanges.

After determining the scale of the fraud on November 3, Crédit Mutuel Arkéa cut external access to its information systems and disconnected from interbank payment systems. It also removed the forged accounts and reset authentication credentials before publicly disclosing the incident the following day. Chairman Julien Carmona said Fortuneo’s systems and customers were not affected.

Carmona said banks and cryptocurrency exchanges had blocked or frozen some transfers and recovered some funds. He estimated the loss at roughly 1% of the group’s equity—equivalent to one quarter’s profit—and said it would not materially affect its operations or financial stability. French police have opened a criminal investigation.

Separately, blockchain investigator ZachXBT identified suspicious cryptocurrency withdrawals before the bank’s disclosure and, on November 4, named Crédit Mutuel Arkéa as a possible victim. Blockchain analytics firm Bitquery traced approximately $108.8 million in USDC leaving accounts associated with Kraken, Crypto.com and OKX over 21 hours on November 1–2; it said a substantial portion was subsequently exchanged for ether. Those tracing figures are not the bank’s final loss figure.

Carmona said the attack very likely involved advanced AI, but offered no evidence or explanation of how it might have been used. No group has claimed responsibility.