Detroit’s use of facial recognition for violent crime probes is down. Here’s why

April 27, 2026

The Detroit Police Department has drastically reduced its use of facial recognition technology, conducting only nine searches in the most recent year compared to over 100 in 2020, representing a 91% decrease from 2023. This reduction follows a 2024 legal settlement stemming from wrongful arrests, which implemented stricter policies requiring independent evidence before arrests can be made based on facial recognition matches. The technology has been particularly criticized for its poor accuracy in identifying Black and Brown individuals, and all nine searches conducted last year involved Black male suspects with only one producing an investigative lead.

Who is affected

Robert Williams, Michael Oliver, and Porcha Woodruff (victims of wrongful arrests due to facial recognition misidentification who filed lawsuits)
Black and Brown individuals (disproportionately affected by the technology's inaccuracy and all nine 2024 searches targeted Black males)
Detroit Police Department (subject to policy restrictions and decreased use of the technology)
Detroit taxpayers (funded over $2 million in facial recognition technology between 2017-2022)
At least 14 people nationwide who have been wrongfully arrested due to facial recognition errors
Local police agencies across Michigan (that access facial recognition through Michigan State Police)
Detroit Board of Police Commissioners (oversight body monitoring the technology's use)
American Civil Liberties Union (legal advocate representing wrongful arrest victims)

What action is being taken

Detroit Police Department is conducting very limited facial recognition searches (only nine in the most recent year)
The department is submitting monthly reports to the Board of Police Commissioners documenting facial recognition use
Detroit police is accessing facial recognition technology through a Memorandum of Understanding with Michigan State Police
Local police agencies across Michigan are submitting facial recognition search requests to Michigan State Police's SNAP Unit
BridgeDetroit is pursuing a Freedom of Information Act request to obtain records about DPD's requests to state police

Why it matters

This matters because facial recognition technology has proven unreliable, particularly in identifying people with darker skin tones, leading to wrongful arrests of innocent individuals and potential violations of civil liberties. The dramatic reduction in Detroit's use demonstrates how policy reforms and legal accountability can substantially change law enforcement practices, potentially serving as a model for other jurisdictions nationwide. The issue has significant implications for criminal justice equity, as the technology disproportionately impacts Black communities, and raises questions about balancing public safety tools with protecting citizens from false accusations. Additionally, the shift away from facial recognition highlights the ongoing debate about whether technological solutions or community-based policing methods are more effective and trustworthy approaches to solving crimes.

What's next

BridgeDetroit is awaiting a response to its narrowed Freedom of Information Act request regarding DPD's facial recognition search requests through Michigan State Police
The ACLU and advocates continue pushing for the complete elimination of facial recognition use, with the goal of reducing searches to zero
An audit of all facial recognition instances since 2017 has been completed by DPD, though the findings summary has not yet been publicly released or provided to BridgeDetroit

Read full article from source: bridgedetroit.com