Accenture Staffer's Claude Agent Ran Two Months on DOD Network

An Accenture staffer connected a personal account for Anthropic's Claude Code to a secure Department of Defense system used by multiple intelligence agencies, and the agent ran on the DOD network for about two months before it was identified, Wired reported.

The agent appears to have sent information about the DOD system, including login credentials, to a high-risk third-party website. The system holds Controlled Unclassified Information, a category that can cover critical infrastructure, nuclear submarine propulsion, and vulnerabilities to chemical and biological terrorism.

An anonymous security source quoted in the piece said the exposure "puts all the data that's in there at risk," and warned that if the third-party software behind the agent were compromised, "there could be somebody out there with the ability to log into a secure environment without anyone knowing." The source called the incident "100 percent a national security threat and could cause irreparable harm."

A DOD spokesperson told Wired that "the Department does not comment on the status of its networks or systems. We take all concerns seriously."

Accenture Federal Services is the integrator the Pentagon picked in July for the Chief Digital and Artificial Intelligence Office's War Data Platform, a $821 million task order covering core integration, cybersecurity, and platform-tool work on the Pentagon's AI data layer. The reporting does not identify which Accenture engagement the staffer was on, which third-party service received the credentials, or how the agent was finally caught. Across the 528 agent stories we have tracked in the past 90 days, this is the first to put a personal-use agent inside a CUI enclave of this sensitivity.