Data Without Due Process: Legal Implications of ICE’s Emerging Data Mining Practices

Abstract

Since 2025, the Department of Homeland Security’s (DHS) Immigration and Customs Enforcement (ICE) has been leveraging AI-driven data mining to analyze and consolidate sensitive medical, employment, biometric, and personal data of individuals across the United States, raising significant legal and privacy concerns. Data mining, the practice of gathering information from large datasets to uncover patterns or extract insights, allows ICE to flag individuals that may be subject to arrest or deportation, based on the information it gathers about them. These practices coincide with an immigration system backlog exceeding three million cases in 2025. This technology, aimed at expediting deportations in the U.S., could further increase the backlog. The rapid spread of AI-driven surveillance, coupled with the absence of clear legal boundaries governing these practices, has heightened concerns about due process and the erosion of privacy protections for both immigrants and non-immigrants. ICE has been able to acquire a vast amount of sensitive information through private contracts with other government agencies, rather than through the legislature. Since these enforcement mechanisms function within limited judicial oversight, they pose significant risks of constitutional violations and unchecked state power. ICE's increased reliance on private data contracts and information gathering exposes a legal loophole: the government may be achieving through private agreements what certain privacy protections would prohibit. This paper argues that ICE’s AI-driven data-mining operations, conducted under vague statutory authority and limited oversight, circumvent Fourth Amendment protections and threaten fundamental privacy rights, demanding urgent legal review and legislative reform.