detroit police car

Opposition to Government Use of Facial Recognition Technology Grows in Detroit

A recent public meeting exposed tensions about how a city-wide video surveillance system is being used to identify crime suspects in Detroit.

Project Greenlight, Detroit’s opt-in video surveillance system for businesses across the city, was launched to much fanfare in 2016. Meant to deter crime through flashing green lights marking the presence of cameras, the public-private initiative has gained support from community leaders, welcoming its 500th business earlier this year.

But the city’s decision to purchase facial recognition software for the camera system in 2017 did not receive a launch event or widespread local media coverage. Up until a few months ago, it was not common knowledge that Detroit police were using the technology and video footage to pursue people accused of crime.

The issue has gained more attention in Detroit and nationally in recent weeks, with The New York Times publishing a feature on police video surveillance and racial bias in the city. The newspaper reported that the facial recognition program matches faces against 50 million driver’s license photographs and mug shots in a statewide database.

May report by the Georgetown Law Center on Privacy & Technology exposed the extent of the real-time facial recognition system, and the police department has since sought a vote from a local oversight board on a policy regarding the issue.

The Georgetown report, titled “America Under Watch,” criticized the city for not being fully transparent with its residents about the program, including the businessowners who signed up for Project Greenlight without knowledge that they were part of a facial recognition operation.

The project’s website also does not explicitly mention the facial recognition system or how police use potential matches to pursue suspects.

“None of the information provided to prospective partners informs them of the fact that face surveillance is part of Project Green Light, and may be used on their camera feeds,” write authors Clare Garvie and Laura Moy.

At a June police commissioners board meeting, residents spoke out against the surveillance program as well as the potential for the city to use facial recognition with a new traffic light program. The board tabled the facial recognition surveillance measure but voted to approve rules for the traffic light cameras, The Detroit Free Press reported.

One local resident, Tawana Petty, told the board that the “very technologists who created [facial recognition] said it’s harmful and biased,” referring to studies that indicate the software makes more errors when it comes to racial minorities and women.

"I have to push you all to at least understand the technology you’re implementing," Petty said, according to the Free Press. "This will be the largest experiment on black people in the United States — 700,000 black people being submitted to an experiment that has already failed all over the world."

San Francisco became the first major American city to ban police use of facial recognition in May despite police and other security officials calling it a useful crime prevention tool.

No such measure has been introduced in Detroit. But even the man tasked with overseeing surveillance technology acknowledged that the software has accuracy issues when it comes to racial minorities. James White, an assistant police chief, said he has been misidentified as other black men by Facebook’s facial recognition algorithm.

“On the question of false positives — that is absolutely factual, and it’s well-documented,” he said, according to the Times. “So that concerns me as an African-American male.”

The Detroit Police Department’s current solution is to limit the use of facial recognition to the investigation of violent crimes, and not arrest anyone solely on the basis of a match, White said.

But some privacy advocates and residents say the potential benefits — catching people who committed crimes without physical evidence — may not be worth the potential harm to innocent people who are misidentified. In their report, Garvie and Moy write that the increased presence of video cameras at schools, religious institutions and community centers under Project Greenlight may violate the privacy rights of citizens and make them less trustful of law enforcement.

“The goal of these surveillance cameras is to make Detroit’s residents feel safe going about their daily lives,” they wrote. “Adding face surveillance to these cameras risks doing the opposite.”

Featured

  • DHS Releases Framework for Safe, Secure Deployment of AI in Critical Infrastructure

    The Department of Homeland Security (DHS) released a set of recommendations for the safe and secure development and deployment of Artificial Intelligence (AI) in critical infrastructure, the “Roles and Responsibilities Framework for Artificial Intelligence in Critical Infrastructure” Read Now

  • Making the Grade with Locks and Door Hardware

    Managing and maintaining locks and door hardware across a school district or university campus is a big responsibility. A building’s security needs to change over time as occupancy and use demands evolve, which can make it even more challenging. Knowing the basics of common door hardware, including locks, panic devices and door closers, can make a difference in daily operations and emergency situations. Read Now

  • Choosing the Right Solution

    Today, there is a strong shift from on-prem installations to cloud or hybrid-cloud deployments. As reported in the 2024 Genetec State of Physical Security report, 66% of end users said they will move to managing or storing more physical security in the cloud over the next two years. Read Now

  • New Report Reveals Top Security Risks for U.S. Retail Chains

    Interface Systems, a provider of security, actionable insights, and purpose-built networks for multi-location businesses, has released its 2024 State of Remote Video Monitoring in Retail Chains report. The detailed study analyzed over 2 million monitoring requests across 4,156 retail locations in the United States from September 2023 to August 2024. Read Now

Featured Cybersecurity

Webinars

New Products

  • A8V MIND

    A8V MIND

    Hexagon’s Geosystems presents a portable version of its Accur8vision detection system. A rugged all-in-one solution, the A8V MIND (Mobile Intrusion Detection) is designed to provide flexible protection of critical outdoor infrastructure and objects. Hexagon’s Accur8vision is a volumetric detection system that employs LiDAR technology to safeguard entire areas. Whenever it detects movement in a specified zone, it automatically differentiates a threat from a nonthreat, and immediately notifies security staff if necessary. Person detection is carried out within a radius of 80 meters from this device. Connected remotely via a portable computer device, it enables remote surveillance and does not depend on security staff patrolling the area. 3

  • Connect ONE’s powerful cloud-hosted management platform provides the means to tailor lockdowns and emergency mass notifications throughout a facility – while simultaneously alerting occupants to hazards or next steps, like evacuation.

    Connect ONE®

    Connect ONE’s powerful cloud-hosted management platform provides the means to tailor lockdowns and emergency mass notifications throughout a facility – while simultaneously alerting occupants to hazards or next steps, like evacuation. 3

  • QCS7230 System-on-Chip (SoC)

    QCS7230 System-on-Chip (SoC)

    The latest Qualcomm® Vision Intelligence Platform offers next-generation smart camera IoT solutions to improve safety and security across enterprises, cities and spaces. The Vision Intelligence Platform was expanded in March 2022 with the introduction of the QCS7230 System-on-Chip (SoC), which delivers superior artificial intelligence (AI) inferencing at the edge. 3