Spain's data regulator received its first breach report blaming an autonomous AI agent
What happened
An organization told Spain's data protection agency that an attacker's AI agent, built on a known language model, logged into its systems, hunted for flaws on its own, altered users' personal data and accessed invoices. The regulator stressed the account is unverified and did not name the organization or the model.
According to the notification, the attacking agent began by scanning generic files for weaknesses, then logged in to the organization's system with working credentials. Once inside, it searched the application for further flaws on its own, found one, and used it to change users' personal data and open invoice and billing records. The agency's deputy director published the account on its blog the day after the notification arrived, stressing that it comes from the affected organization and has not yet been analysed. He also noted that the use of a particular AI model does not mean the model's provider was compromised or that the tool was built for attacks. The regulator has not said how long the agent had access, how many records were altered, how the credentials were obtained, or whether data left the system.
How it happened
AI-run attack: Autonomous intrusion. The attacker's own AI agent carried out the break-in, working through the steps with little human involvement.
Sources do not say whose AI model was involved. The product type was AI agent.
Impact
Data was both exposed and destroyed. Data involved: personal data and financial. The impact was confined to one organization and its users or customers.
Severity is rated High, calculated from how far the incident reached and how sensitive the data was.
Was the AI at fault?
No. The failure was in the systems, settings or people around the AI, not in the model's behaviour.
Status and timeline
Reported in the press or by researchers, but not confirmed by the affected organization, the AI vendor or a regulator.
It became public on 15 Sep 2026.
The AEPD urged organizations to add AI-executed attacks to their risk analyses, shorten response times, limit access and fix known vulnerabilities, and reminded them that breaches must still be reported within 72 hours under GDPR regardless of what carried them out.