Keep Factually independent
Whether you agree or disagree with our analysis, these conversations matter for democracy. We don't take money from political groups - even a $5 donation helps us keep it that way.
What databases and data-sharing agreements does ICE use to identify immigration enforcement targets?
Executive summary
ICE relies on a mix of internal DHS databases (like the Enforcement Integrated Database/EID) and a web of interagency memoranda and agreements—most prominently a 2025 IRS-ICE memorandum of understanding (MOU) that would let ICE request taxpayer names, addresses and certain tax data to locate people with final removal orders or under criminal investigation (court fights and injunctions have limited that sharing) [1] [2] [3]. Advocates and courts say the IRS MOU is unprecedented and legally contested; civil‑liberties groups warn it could chill tax filing, while some courts have halted or narrowed the practice [4] [5] [6] [3].
1. Internal DHS systems: the backbone — EID and related tools
ICE and other DHS components use the Enforcement Integrated Database (EID) as a shared repository for arrest, detention, removal and investigative records; EID supports interfaces and tools (EAGLE, EDDIE, DAVID) that process personally identifiable information about people ICE encounters or seeks to remove [1]. The DHS privacy impact assessment for EID documents how the database aggregates investigative and biometric data and how multiple DHS applications query that information, making it a primary internal source ICE uses to identify targets [1].
2. Interagency accords and MOUs: IRS-ICE as a recent flashpoint
In 2025 the IRS and DHS/ICE formalized an MOU that allowed ICE to request taxpayer names, addresses and certain tax-period information to aid enforcement against people with final removal orders or subject to criminal investigations; the MOU’s existence and scope spurred resignations, litigation and public debate [2] [7] [4]. Courts initially allowed parts of the arrangement to proceed but later federal rulings paused or narrowed data transfers after judges questioned the breadth of requests and whether the IRS complied with statutory limits on tax-data sharing [8] [3] [6].
3. How the IRS sharing was intended to work — and why critics object
Explainers and legal analyses describe the mechanism as ICE submitting lists (names/addresses/periods) that the IRS would match to administrative records and verify addresses or returns; proponents said it would help locate people subject to removal, while critics call it an unprecedented repurposing of confidential tax data that could discourage tax compliance and sweep in millions of taxpayers [9] [4] [10]. Civil‑liberties groups and privacy advocates framed the MOU as weaponizing taxpayer trust and warned of large downstream social and fiscal costs if people stop filing taxes [5] [4].
4. Legal pushback, injunctions and operational limits
Multiple organizations sued to block IRS disclosure; some courts initially allowed the agreement to operate but later courts issued injunctions halting transfers and criticizing the breadth of the IRS’s plan to hand over address data to a small number of ICE officials—finding that statute and practice limit IRS disclosure to officials directly involved in a discrete, ongoing investigation [8] [3] [6]. Reporting indicates DHS sought large name lists (about 1.2–1.3 million) and that tens of thousands of matches were returned before courts intervened, underscoring how operational-scale sharing had already begun [6] [3].
5. Other data flows and local cooperation highlighted by advocates
Advocacy groups and watchdogs document a broader “immigration enforcement web”: ICE receives information from local law‑enforcement partners, probation/parole officers, and other agencies, and uses information‑sharing agreements and access handbooks to formalize these links; NILC and similar groups map how those routines feed ICE identification and arrest actions [11] [12]. This perspective emphasizes that ICE’s targeting is not only driven by federal databases but also by state/local cooperation and routine data exchanges [11] [12].
6. Emerging technologies and public controversy
Reporting has flagged specific tools—such as a mobile facial‑recognition application (reported as “Mobile Fortify”) that reportedly queries multiple federal databases to identify people encountered in public—and lawmakers and civil‑liberties groups have pushed for pauses or transparency about such technology use [13]. Independent pieces claiming AI “Hurricane Score” tools exist appear in secondary outlets, but available sources do not mention authoritative government documentation describing that specific AI score in official ICE materials (p1_s7; available sources do not mention the “Hurricane Score” in [1] or official PIAs).
7. What reporting does not (yet) settle
Available sources document EID, the IRS-ICE MOU and judicial actions, plus advocacy analyses of local data flows [1] [2] [3] [11]. However, available sources do not mention a comprehensive public inventory listing every external database ICE queries beyond EID and the IRS MOU, nor do they provide exhaustive technical specifics about every tool’s data sources and matching algorithms—courts, redactions and “law‑enforcement sensitive” designations have kept parts of the playbook opaque (p2_s10; available sources do not mention a full public inventory).
Bottom line: the public record shows ICE’s core identifiers come from DHS systems like EID and an array of interagency agreements, with the 2025 IRS‑ICE MOU a prominent, legally contested example of how non‑immigration federal data (tax records) was being weaponized for immigration enforcement until courts intervened [1] [2] [3].