Revolut Confirms Data Breach Involving Unauthorized Disclosure of Customer Information
British fintech company Revolut has reported a serious data breach, revealing that it mistakenly shared personal information of a small, unspecified number of customers with an unauthorized third party. This breach was facilitated by fraudulent information requests sent from the email domain of a legitimate governmental agency. In response, the company has alerted the impersonated agency, police forces, financial regulators, and data protection authorities.
Throughout the year, banks, telecommunications operators, and major platforms regularly respond to administrative and judicial requisitions. This continuous flow primarily aligns with compliance rather than cybersecurity, dealt with the discretion demanded by ongoing investigations. Few companies have approached this experience as a potential attack vector. On September 12, 2026, Revolut disclosed an incident illustrating the risks when an attacker successfully infiltrates such processes.
A spokesperson for the digital bank described the incident as a “sophisticated external impersonation scam,” where an unauthorized third party utilized an official government agency email domain to submit fraudulent information requests. According to the statement referenced by Reuters on September 12, “Upon detection, we immediately blocked the email address and notified the relevant government agency, law enforcement, data protection authorities, and financial regulators.” Affected customers were informed individually, while Revolut has refrained from naming the impersonated agency.
An Authenticated Channel, A False Request
SPF, DKIM, and DMARC protocols successfully verified this email as legitimate since it came from the authentic email domain of the agency rather than a spoofed one. These three protocols serve their intended purpose: to confirm that a message originates from the domain it claims to represent. However, they remain silent on the legitimacy of the request contained within. An authenticated channel, even when properly implemented, does not warrant consent for the request being processed. The lack of control at the time of the incident is suggested by the timeline, indicating a need for prior verification of the sender outside of the communication channel.
Since 2022, various published investigations have established that attackers have been acquiring data from major platforms by submitting false emergency requests from compromised police service email accounts. This blind spot presents similar vulnerabilities for financial institutions. Data breaches reported in France total in the thousands each year, with no distinction made for those involving indirect procedures.
What Revolut Confirms, What Attackers Claim
Revolut confirms that data was disclosed via fraudulent requests and asserts that its information systems and customer funds remain secure. While this latter point is indeed valid, it answers a question that had not been raised. A lost passport copy can still hold value for identity theft long after a system has been secured.
On the same day, TechCrunch published a summary of the transmitted data, which Reuters also covered, although no inventory data has been provided by Revolut. The reported information includes birth dates, postal and email addresses, phone numbers, and copies of identification documents like passports and driver’s licenses. Digital asset press outlets have also mentioned selfie verification, account statements, and transaction histories, the extent of which the company has not confirmed at this stage.
Furthermore, malicious actors have opened a distribution channel where they have been posting customer knowledge documents since September 13, claiming to possess a broader database obtained from another source. However, this claim remains unverified, as neither Revolut nor any independent source has substantiated it. The assertion is presented as such and should not be treated as fact.
Revolut describes the number of affected customers as “very limited” without providing specifics, and no authority has released estimates as of now. The incident occurs as the company is preparing for an initial public offering potentially valuing it at up to $200 billion, according to Reuters.
Administrative Requisition: An Untapped Attack Surface
Tax authorities, investigative services, and sector regulators submit their requests via email, often accompanied by tight deadlines and confidentiality directives. These requests are routed to a lawyer or compliance officer, bypassing the access controls that IT departments enforce for their systems. Every exposed company receives such inquiries.
In response to this lack of authorization, the General Data Protection Regulation (GDPR) obligates the data controller to establish what information has been shared, with whom, and on what basis. This raises four key questions: Who specifically has permission to respond to authority requests? What independent verification of the source was conducted, outside of the incoming channel? What thresholds were used, given that a passport copy doesn’t represent the same level of verification as a mere account existence confirmation? Finally, what documentation accompanies this process, considering that verification is done post facto?
An AI agent connected to the compliance inbox to prepare responses would shift the point of human control—the very point that ultimately detected the fraud at Revolut. Today, a guarantee to take its place remains elusive, with agents gaining access to identity systems faster than organizations learn to govern them. The confidentiality of investigations, invoked in many of these requests, complicates verification and weighs heavily on compliance teams, yet it still allows for sender verification, a process focusing on the origin of the request while keeping its content intact.
This verification process remains the only barrier. A company disclosing personal data in response to a falsified request retains responsibility as a data controller under GDPR, regardless of the quality of the deception. Currently, no active European legislation, including the NIS2 directive or the AI regulation, establishes a protocol for verifying authority requests. Data protection authorities are expected to clarify in the coming weeks what level of diligence they expect for this channel, a response that will undoubtedly impact sectors beyond financial services.



