Milwaukee Police Chief to ban department's use of facial recognition technology – WISN

A Landmark Decision: Milwaukee Police Department Halts Use of Controversial Technology

In a significant policy shift that places the city at the forefront of a contentious national debate, Milwaukee Police Chief Jeffrey Norman has announced a department-wide ban on the use of facial recognition technology. The decision marks a definitive stance against a powerful surveillance tool that has drawn intense scrutiny from civil liberties advocates, technology experts, and community leaders over concerns about its accuracy, potential for racial bias, and profound implications for personal privacy.

The move represents a deliberate step back from the technological frontier of policing, prioritizing community trust and constitutional rights over the unproven and ethically fraught benefits of an algorithm-driven identification system. While law enforcement agencies across the country have explored or adopted facial recognition as a means to generate leads and identify suspects, a growing chorus of critics has warned that the technology is a gateway to mass surveillance and can perpetuate and even amplify existing inequalities within the criminal justice system.

Chief Norman’s directive effectively sidelines the technology as an investigative tool for Milwaukee’s officers, signaling a fundamental reassessment of the balance between public safety and individual freedom. This decision is not merely a technical or procedural adjustment; it is a statement of principle that will reverberate through city hall, community forums, and police departments nationwide that are grappling with the same complex questions. As Milwaukee charts a course away from automated facial analysis, the department is implicitly championing a model of policing rooted in human-led investigation and a commitment to equitable and transparent practices.

Understanding Facial Recognition: The Technology at the Center of the Debate

To fully grasp the magnitude of Milwaukee’s decision, it is essential to understand what facial recognition technology is, how it functions, and why it has become one of the most polarizing innovations in modern law enforcement.

How Facial Recognition Technology Works

At its core, facial recognition technology (FRT) is a form of biometric surveillance. The process begins with capturing a face from an image or video. Specialized software then analyzes the geometry of the face, measuring key features and landmarks known as nodal points. These include the distance between the eyes, the depth of the eye sockets, the shape of the cheekbones, and the length of the jawline. This data is converted into a unique numerical code, or “faceprint,” which is essentially a digital signature for an individual’s face.

This newly created faceprint is then compared against a vast database of existing faceprints. The database can be compiled from a variety of sources, including government-held images like mugshots and driver’s license photos, or, more controversially, from publicly available images scraped from social media and other websites. The algorithm searches for potential matches by calculating the similarity between the probe faceprint and the millions of faceprints in its database, ultimately producing a ranked list of possible candidates.

Applications in Modern Law Enforcement

Police departments typically use FRT not as a definitive means of identification but as an investigative lead generator. It is crucial to note that the technology does not, on its own, identify a criminal. Instead, it offers a “possible match” that detectives must then corroborate through traditional police work, such as interviewing witnesses, collecting further evidence, and confirming the suspect’s whereabouts.

Common use cases include:

• Identifying Unknown Suspects: An officer might feed a still image from a grainy convenience store surveillance video into the system to generate a list of potential suspects.
• Identifying Uncooperative Individuals: If a person in custody is uncooperative or provides false identification, FRT might be used to determine their true identity by matching their face to a mugshot database.
• Finding Missing Persons or Victims: The technology can be used to identify victims of crime who are incapacitated or deceased, or to help locate missing children or vulnerable adults.

Proponents argue that when used responsibly, FRT is a powerful tool that can accelerate investigations, solve difficult cases, and bring perpetrators of serious crimes, including human trafficking and child exploitation, to justice.

The Rise of Third-Party Vendors and “Scraped” Data

The proliferation of FRT in law enforcement has been fueled by private technology companies that offer “policing-as-a-service” solutions. Firms like the controversial Clearview AI have amassed colossal databases by scraping billions of images from the public internet—including social media platforms like Facebook, Instagram, and YouTube. This practice has raised significant ethical and legal alarms, as these images are often harvested without the knowledge or consent of the individuals pictured.

The use of these third-party systems introduces a layer of opacity. The algorithms are proprietary, functioning as “black boxes” whose inner workings are not open to public or independent scrutiny. This makes it impossible to fully assess their accuracy, biases, or error rates, forcing police departments and the public to simply trust the vendor’s claims—a proposition that has become increasingly untenable.

The Core of the Controversy: Why a Ban Was Implemented

The decision by the Milwaukee Police Department to ban facial recognition technology is a direct response to a trifecta of deeply troubling issues that have been documented by researchers, legal scholars, and civil rights organizations.

The Specter of Systemic Racial Bias

Perhaps the most damning criticism of FRT is its well-documented racial and gender bias. Groundbreaking studies, including a landmark 2019 report from the federal government’s National Institute of Standards and Technology (NIST), have found that the majority of facial recognition algorithms exhibit significant demographic disparities. Specifically, these systems are far more likely to produce false positives—incorrectly matching a face to an unrelated person in a database—when analyzing the faces of African Americans, Asian Americans, and women.

The NIST study found that some algorithms were up to 100 times more likely to misidentify a Black or East Asian face compared to a white male face. This bias is largely attributed to the data used to train the algorithms. When these systems are predominantly “taught” using images of white men, they become less adept at distinguishing the unique facial features of other demographic groups.

In the context of the American criminal justice system, which already faces persistent challenges with racial disparity, introducing a biased technology risks compounding injustice. An erroneous match could place an innocent person of color under suspicion, leading to a wrongful arrest, financial hardship, and lasting trauma. The fear is that FRT could become a high-tech pipeline for reinforcing and accelerating pre-existing biases in policing.

Accuracy, Reliability, and the High Stakes of Misidentification

Beyond demographic bias, the overall accuracy of the technology remains a serious concern, especially when used in real-world conditions. Factors like poor lighting, low-resolution images, odd camera angles, or partial facial occlusions (e.g., from a hat or sunglasses) can dramatically increase error rates. A system that performs well in a controlled lab setting may fail spectacularly when analyzing blurry, real-world surveillance footage.

The consequences of these failures are not abstract. Across the United States, there have been multiple documented cases of innocent people being wrongfully arrested based on faulty facial recognition matches. In 2020, Robert Williams, an African American man from Michigan, was arrested in front of his family and held for 30 hours for a theft he did not commit, all due to a flawed algorithm. Similar cases, like that of Nijeer Parks in New Jersey, have highlighted the devastating human cost of relying on an imperfect technology for something as critical as determining a person’s guilt or innocence.

These high-profile failures have served as a stark warning to law enforcement leaders, demonstrating that an over-reliance on FRT can lead to catastrophic miscarriages of justice that undermine the very legitimacy of the police.

Erosion of Privacy and the Chilling Effect on Civil Liberties

The third major pillar of opposition to FRT is its profound threat to privacy and fundamental civil liberties. The ability to identify individuals in real time from a distance, and to do so on a mass scale, fundamentally alters the relationship between the citizenry and the state. Civil liberties groups argue that the widespread deployment of FRT creates a “perpetual lineup,” where every person in a public space is subject to constant, passive identification and tracking.

This capability raises fears of a future where governments could monitor attendance at political protests, religious services, or community meetings, creating a detailed record of people’s associations and movements. The knowledge that one is being watched can create a powerful “chilling effect,” deterring people from exercising their First Amendment rights to free speech and assembly for fear of being misidentified or unfairly targeted by authorities.

Furthermore, the creation of massive, centralized facial databases—whether government-run or privately held—presents a tantalizing target for hackers and a potential tool for misuse. A data breach could expose the biometric information of millions of people, a form of identity theft that is impossible to remedy—unlike a stolen password, a person cannot simply get a new face.

Milwaukee in the National Context: A Growing Movement of Skepticism

The Milwaukee Police Department’s ban is not an isolated event. Rather, it aligns the city with a growing coalition of municipalities and states that have decided the risks of facial recognition technology outweigh its purported benefits.

A Patchwork of Policies: Cities and States Take the Lead

In the absence of comprehensive federal regulation, a patchwork of local and state laws has emerged. Major cities like San Francisco, Boston, and Portland (both Oregon and Maine) have all passed outright bans on government use of the technology. Dozens of other municipalities have implemented moratoriums or strict regulations governing its use.

At the state level, states like Illinois have long-standing biometric privacy laws that create high hurdles for the use of FRT, while Washington and Virginia have passed laws that seek to regulate, rather than ban, its use by police. This fragmented landscape highlights the lack of national consensus and the deep divisions over the technology’s role in a democratic society.

By joining the ranks of cities that have implemented a full ban, Milwaukee is making a clear statement that reform and regulation are insufficient to address the technology’s fundamental flaws. This positions the MPD as a leader in a movement that prioritizes a precautionary approach, arguing that such a powerful surveillance tool should not be deployed until it can be proven safe, accurate, and equitable.

The Counterargument: Proponents Argue for Regulation, Not Prohibition

It is important to acknowledge that the push for bans is not universally supported. Many within the law enforcement community and the tech industry argue that prohibiting facial recognition is a mistake that deprives police of a valuable tool for ensuring public safety. They contend that, like DNA or fingerprint analysis, FRT can be a game-changer for solving heinous crimes and exonerating the innocent.

The argument from this perspective is not to abandon the technology, but to regulate it. Proponents of this approach advocate for policies that would mandate transparency in how the technology is used, require judicial oversight, implement strict accuracy standards, and prohibit its use for monitoring First Amendment-protected activities. They believe that a framework of robust checks and balances can mitigate the risks while preserving the benefits. The decision in Milwaukee, however, reflects a deep skepticism that such a framework can ever be truly effective against a technology with such a high potential for abuse and error.

The Path Forward: Policing in a Post-Facial Recognition Era in Milwaukee

With facial recognition technology off the table, the Milwaukee Police Department is now poised to redefine its approach to modern investigation, with a renewed emphasis on foundational principles of police work.

A Renewed Focus on Rebuilding Community Trust

Chief Norman’s decision is, in many ways, an act of trust-building. By proactively removing a tool that is a source of significant fear and suspicion, particularly in communities of color, the department is sending a message that it is listening to community concerns. This move can be leveraged to open new dialogues and strengthen relationships between officers and the residents they serve.

The path forward will likely involve a recommitment to community-oriented policing strategies that rely on human intelligence, collaboration, and mutual respect. Building trust is not a technological problem; it is a human one. By stepping away from a controversial piece of tech, the MPD creates an opportunity to invest its resources and energy into programs that foster positive, non-enforcement interactions and collaborative problem-solving with the community.

Emphasizing Traditional and Alternative Investigative Methods

The ban on FRT necessitates a reliance on and enhancement of other investigative techniques. This includes:

• Forensic Evidence: A continued focus on the meticulous collection and analysis of physical evidence, such as DNA and fingerprints, which are subject to more established scientific and legal standards.
• Witness and Victim Interviews: Doubling down on the core police skill of gathering information through direct human interaction and thorough interviewing.
• Public Engagement: Leveraging the community as a resource by publicly disseminating images of suspects and asking for tips—a transparent, consent-based approach to identification.
• Digital Forensics: Analyzing other forms of digital evidence, such as cell phone location data or social media activity, under the authority of a warrant.

While the absence of FRT may make identifying a suspect from a photo more challenging in some cases, the department is betting that a focus on these proven, less controversial methods will yield just outcomes without compromising the civil rights of Milwaukee’s citizens.

A Defining Moment: The Broader Implications of Milwaukee’s Stance

The Milwaukee Police Department’s decision to ban facial recognition technology is more than a local policy change; it is a defining moment in the ongoing struggle to reconcile technological advancement with the core values of a free and equitable society. It represents a conscious choice to prioritize human rights over algorithmic efficiency and to affirm that the legitimacy of policing is built on a foundation of trust, not surveillance.

This bold move forces a critical and necessary conversation. It challenges the notion that police departments must adopt every new technology available to them and instead asks a more fundamental question: What kind of policing do we want? In Milwaukee, the answer provided by this ban is clear: a form of policing that is transparent, accountable, and steadfast in its protection of the constitutional rights of every individual it serves.

As other cities across the nation continue to debate the future of surveillance, they will now look to Milwaukee not as a city that fell behind the technological curve, but as one that stepped forward to lead on the principles of justice.