AI agents ran an enterprise intrusion end to end in under ten hours, Unit 42 says
What happened
Unit 42 investigated an intrusion in which attackers used frontier models and agentic frameworks for every stage, taking hard-coded secrets, administrative credentials and cloud keys from an enterprise network in under ten hours. An agent left the victim an 80-page report on its security failings.
Unit 42 says the attackers accessed source code repositories and secrets, pulled master administrative credentials, stole cloud access keys and used the victim's cloud AI endpoints for their own compute. It estimates the work would take human operators about two weeks. A 3 September update clarifies it was an intrusion, not ransomware.
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 exposed. Data involved: credentials and source code. 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?
Unclear. Sources do not establish whether the model's own behaviour or the surrounding systems were to blame.
Status and timeline
Confirmed by the affected organization, the AI vendor, a regulator or a named security research firm.
It became public on 2 Sep 2026.