.png "Global Insight")
One of the largest leaks of US government personnel data in years landed on October 28, when a transparency group published a searchable database covering 41,191 employees of the Department of Homeland Security — including people working in undercover operations. According to reports from Reuters and NPR, the dataset, published by Distributed Denial of Secrets under the name "Icepick", lists names, phone numbers, email addresses, office locations, job titles and internal group memberships across nine DHS agencies.
What exactly leaked
The scale is what makes this leak different from an ordinary breach. Roughly 21,000 of the people listed work for Customs and Border Protection and about 6,600 for Immigration and Customs Enforcement, with the rest spread across Homeland Security Investigations, the Federal Protective Service, the Cybersecurity and Infrastructure Security Agency and other components. The records span nearly 7,000 offices and more than 2,200 locations. What worries security officials most is not just the contact details but the group memberships: they reveal which agents are assigned to special operations, surveillance, computer forensics and undercover work, and they expose operation names and surveillance programmes that were previously unknown. Reports describing the data also mention groups tied to surveillance contractors and a facial-recognition app used by agents in the street.
The material was provided by a hacktivist group calling itself the Department of Peace, which says it obtained the records by subverting DHS's single sign-on system — the system staff use to log in to internal services. The group shared the data with Distributed Denial of Secrets and with a Swiss hacker and journalist known as maia arson crimew, who built the public search tool. One analysis of the data noted timestamps pointing to mid-August, raising the question of how long the department may have had a compromised staff directory before the leak became public — a question DHS has not yet answered.
The dispute over how it happened
Not everyone agrees on the mechanism. The hacktivists point to Okta, the single sign-on provider whose service DHS uses; Okta has said it has no evidence its data or systems were compromised. That dispute matters because it determines the fix: if the flaw was in how DHS configured its login system, other agencies using similar setups need to check their own configurations; if a vendor was breached, the problem is wider still. Reuters matched six named employees in the leak against credit-bureau and breach records and telephoned three of them, who confirmed their details — one said it was the first they had heard of any breach. In a statement, DHS said it was aware of a potential breach and was working to address it swiftly, adding that it takes threats against its employees seriously and will work to ensure they are properly protected.
Accountability versus safety
The leak's publishers frame it as accountability: Distributed Denial of Secrets co-founder Emma Best said the tool would replace anonymity with accountability and help journalists, public-records requests and congressional oversight. Critics — and DHS itself — frame it as a safety risk: undercover agents whose identities are exposed can face retaliation, and criminal networks can use organisational charts to map investigations. Both arguments have force, and the tension between them is the real story here. This is the third major US government data leak in about a month, after a September breach claimed against FBI job portals and a Pentagon breach affecting more than 3 million people — a pattern that suggests federal personnel systems, not just one department, are under sustained pressure. As with earlier warnings about US security vulnerabilities and the strain on US-linked security operations, the question now is less whether more leaks will come than whether agencies can protect the people whose names are already out.