DOJ ramps up AI for legal work, crime predictions, surveillance, inventory shows (fedscoop.com)

The Department of Justice (DOJ) has reported a significant expansion in its use of artificial intelligence (AI) across various functions, increasing from four identified use cases two years ago to 315 in 2025—an increase of 31% from last year. This comprehensive inventory highlights AI applications ranging from crime prediction and public surveillance to improving litigation processes. Notably, 114 of these cases were classified as “high-impact,” suggesting they may have profound effects on legal rights and safety. The DOJ's increasing reliance on predictive AI tools, such as those assessing inmate recidivism and behavior, has raised concerns about inherent biases and ethical implications, particularly regarding the potential for discriminatory outcomes. As the DOJ leverages AI for tasks like analyzing travel data in criminal investigations and streamlining legal workflows, experts warn of the implications this technology might have on civil liberties and power dynamics within the justice system. In particular, the use of facial recognition technology, including Amazon Rekognition, continues to evoke privacy concerns as it is being considered for extensive surveillance purposes within the DOJ. Critics argue that as these technologies become more embedded in law enforcement practices, the risk of bias and the erosion of privacy rights intensifies, drawing attention to the urgent need for accountability and transparency in AI deployments within the federal agency.