Surveillance Cities

Privacy & Surveillance

Cities & Surveillance: From CCTV to Flock

Two decades ago, the debate was about a few hundred street cameras. Today it’s license-plate readers on nearly every corner, doorbell-camera networks, and facial recognition. Here’s how municipal surveillance evolved, what the research says about whether it works, and why it’s back at the center of public debate.

Reviewed & updated · 10 min read

Surveillance Today, by the Numbers

The scale has changed dramatically since the early CCTV pilots of the 2000s.

~85M
Cameras in the U.S.
Estimated installed nationwide (industry estimates)
5,000+
Flock ALPR communities
Across 49 states, ~20 billion scans/month (2025)
99.9%
Of ALPR data is innocent
Unconnected to any crime, per EFF
$4→$1
Best-case ROI
Chicago’s savings per dollar spent (Urban Institute)

The Modern Surveillance Landscape

License-plate readers & Flock Safety

The biggest shift is the rise of automated license-plate readers (ALPRs). Flock Safety, the best-known vendor, says its cameras operate in more than 5,000 communities across 49 states and perform on the order of 20 billion vehicle scans a month. ALPRs photograph every passing car — logging plate, location, time, and distinguishing features — and the EFF estimates more than 99.9% of that data has no connection to any crime. In 2025 the technology drew intense scrutiny after reports that local ALPR data was queried for federal immigration enforcement, sometimes in violation of state law; since early 2025, at least 30 localities have deactivated their Flock cameras or canceled contracts.

Doorbell cameras & Ring

Consumer doorbell cameras created a vast private surveillance network that police could tap. In January 2024, after years of criticism, Amazon’s Ring ended the “Request for Assistance” tool that let police solicit user footage through its Neighbors app — though law enforcement can still obtain footage with a warrant or subpoena.

Facial recognition

Facial recognition has become powerful and contentious. At least 21 cities and counties across 11 states, plus the state of Vermont, have restricted police use, and there have been multiple documented wrongful arrests tied to facial-recognition errors — nearly all involving Black Americans. In 2024, the firm Clearview AI was fined €30.5 million by the Dutch data-protection authority over its face database. Accuracy, bias, and oversight remain central concerns.

Do Cameras Prevent Crime? Early Case Studies

A landmark Urban Institute study examined public-camera systems in three cities. The finding: where cameras were sufficiently concentrated and actively monitored, the impact on crime was significant and cost-effective, with no evidence that crime simply moved elsewhere.

Baltimore, MD

Grew from a 5-camera pilot to 500+ downtown and in high-crime neighborhoods, monitored by retired officers. The Urban Institute found significant declines in violent crime and larceny (2003–2008), with benefits extending beyond camera view and no displacement — though some areas (e.g., North Avenue) saw no change.

Chicago, IL

Built one of the largest networks — thousands of “police observation devices” (PODs) with flashing blue lights and real-time feeds. Results were mixed by neighborhood, but the program was credited with major monthly criminal-justice savings; the Urban Institute estimated Chicago saved about $4 for every $1 spent.

Washington, D.C.

Installed cameras from 2002 and expanded after a 2006 homicide spike, with strict guidelines and centralized monitoring. Results were inconsistent — some violent crime fell while theft rose — but officials valued the system for investigations and public reassurance.

Source: Urban Institute, Evaluating the Use of Public Surveillance Cameras for Crime Control and Prevention. The U.S. National Institute of Justice rates CCTV as “Promising” for property and vehicle crime.

The Privacy Debate

The case for

Supporters say cameras and ALPRs deter offenders, alert police to crimes in progress, and provide evidence to identify suspects, vehicles, and witnesses. Research shows real, cost-effective crime reductions where systems are dense and actively monitored — and footage routinely aids investigations and prosecutions.

The case against

Civil-liberties groups warn that mass collection — where 99.9% of data is innocent — builds detailed portraits of ordinary people, can chill protected activity, and is prone to misuse, data-sharing, and error (including wrongful arrests). Effectiveness also varies widely by location and design, raising cost-vs-benefit questions.

Resources & Research

Authoritative, current sources on municipal surveillance — from research bodies and civil-liberties groups.

Map

EFF Atlas of Surveillance

A searchable map of police surveillance tech — ALPRs, cameras, drones, face recognition — by U.S. locality.

Visit →
ALPRs

EFF: License-Plate Readers

How ALPRs work, what data they collect, and the privacy concerns — from EFF’s Street-Level Surveillance project.

Visit →
Civil Liberties

ACLU: Surveillance Technologies

The ACLU’s analysis of cameras, face recognition, and location tracking, and the policies around them.

Visit →
Research

Urban Institute Study

The benchmark evaluation of public cameras in Baltimore, Chicago, and Washington, D.C.

Visit →
Evidence Rating

NIJ CrimeSolutions: CCTV

The National Institute of Justice’s evidence rating of CCTV surveillance for crime prevention.

Visit →
Doorbell Cameras

EFF: Ring & Police (2024)

How and why Amazon’s Ring ended its police footage-request tool, and what it means.

Visit →
Responsible Video Surveillance

Cameras for Your Property, Done Right

JMAC Supply distributes IP cameras, recorders, and access-control systems for homes, businesses, and facilities — the kind of focused, on-premises security that protects your own property. Learn more in our history of CCTV and alarm monitoring guides.

Frequently Asked Questions

Municipal surveillance, explained.

An ALPR is a camera system that automatically photographs passing vehicles and reads their license plates, recording the plate, time, and location. Networks like Flock Safety aggregate these reads across thousands of communities. Civil-liberties groups note the vast majority of the data collected has no connection to any crime.

Industry estimates put the number around 85 million cameras installed across the United States, spanning homes, businesses, and public spaces. Camera density varies widely by city, with some metros having well over 100 cameras per 1,000 residents.

Sometimes — it depends on design. The Urban Institute found significant, cost-effective crime reductions where cameras were concentrated and actively monitored (Chicago saved roughly $4 per $1 spent), with no displacement, but little effect where coverage was thin or unmonitored. The NIJ rates CCTV “Promising” for property and vehicle crime.

As of January 2024, Ring no longer lets police request footage directly through its Neighbors app. However, law enforcement can still obtain footage with a warrant or subpoena, and you can choose to share clips voluntarily.

Not reliably. There have been several documented wrongful arrests in the U.S. tied to facial-recognition errors — nearly all involving Black individuals — and studies have found higher error rates for some demographic groups. That’s why at least 21 cities and counties across 11 states, plus Vermont, have restricted police use.

Securing Your Own Property?

JMAC Supply has the cameras, recorders, and access-control hardware to protect your home or business — plus the expertise to design it well.

Live Chat