The New York City Council Committee of the Whole held a hearing on October 5, 2026, about the risks posed by artificial intelligence (AI) and potential frameworks for AI oversight. The hearing featured testimony from Anthropic, OpenAI, Google, and Meta executives, public officials, civil society, and New York City residents. Rachel Levinson-Waldman, director of the Liberty and National Security Program, testified before the Committee to raise concerns regarding the use of AI by the New York City Police Department (NYPD), warning that existing oversight and disclosure mechanisms are failing to protect New Yorkers from surveillance overreach and are not adequately safeguarding their civil rights and civil liberties.
The NYPD has confirmed use of AI across various surveillance technologies, including facial recognition, gunshot detection, and social media monitoring software, through public testimony and Impact and Use Policies mandated under the Public Oversight of Surveillance Technology (POST) Act. However, these disclosures fail to provide details about how AI is integrated in the Domain Awareness System (DAS) — the Department’s data-fusion system that aggregates information collected across surveillance technologies. As Levinson-Waldman emphasized in her testimony, the Council must enforce existing oversight mechanisms to increase transparency and governance of the use of AI within the NYPD.
Levinson-Waldman identified several reforms that the Council should adopt to protect New York City residents from the risks posed by the NYPD’s use of AI. First, the Council must ensure that the Office of Inspector General for the NYPD has the funding, permanent leadership, and technical expertise it needs to scrutinize the NYPD’s use of AI. To operate effectively, the Department of Investigations must appoint a new Inspector General and receive the necessary support to increase staffing levels. Second, the Council must require the Office of Technology and Innovation to fulfill its obligation to fully staff the Office of Algorithmic Accountability (OAA), which was supposed to open its doors in June 2026, and does not yet have a permanent director or other staff. Lastly, the Council should hold a hearing dedicated to the NYPD’s use of AI where the Department is required to explain what AI-enabled tools it uses, how AI influences policing decisions, and what safeguards protect New Yorkers.
The Brennan Center has continually advocated for increased oversight of AI use by the NYPD. In 2023, then-Brennan Center counsel Ivey Dyson co-wrote an article with New York City Council Member Jennifer Gutiérrez calling for greater oversight of the NYPD’s surveillance practices. In 2024, we published reports documenting the inadequacy of surveillance oversight in New York City and Washington, D.C.. And last year, Levinson-Waldman and Dyson published an expert brief examining the harms of unregulated AI-powered policing tools, including data fusion platforms such as DAS.
