A growing, bipartisan backlash against security-technology company Flock Safety is emerging in communities across the United States. Critics argue that the Georgia-based company’s extensive network of automated license plate readers (ALPRs) gathers sensitive information in ways that may conflict with Fourth Amendment protections. Flock maintains that its technology enhances public safety and operates within the law. However, a series of errors and incidents has raised concerns about how the system can be misused.
Although Flock has recently become a prominent name in the national debate over surveillance, the controversy represents a much broader problem: the widening gap between constitutional privacy protections and the realities of increasingly sophisticated digital surveillance.
Flock cameras are operating across thousands of American communities
The use of surveillance cameras in public spaces is hardly new. What has changed is the rapid development of artificial intelligence, which has given Flock and similar companies the ability to collect, analyze and track information on a scale that has alarmed civil-rights organizations.
Flock reportedly has about 120,000 automated license plate readers deployed in roughly 7,000 communities throughout the United States. These cameras record license plates as well as identifying characteristics of vehicles, including their color, bumper stickers and visible damage. The information can then be used by law enforcement to locate and track vehicles. Newer versions of the technology have also been shown to identify and follow pedestrians.
Because the cameras operate continuously, they inevitably record the movements of large numbers of people who are not suspected of any wrongdoing. Flock argues that this does not amount to a Fourth Amendment violation. The Fourth Amendment, part of the U.S. Bill of Rights, generally restricts government searches and has been interpreted by courts to provide protection for certain forms of digital location information, including cellphone-location records, when obtained without appropriate legal authorization.
Flock, as a private company, is not itself governed by exactly the same constitutional restrictions as government agencies. Nevertheless, its business model is closely tied to law-enforcement agencies that use its technology. That relationship has intensified concerns among civil-rights advocates about privacy, accountability and the potential consequences when automated surveillance systems make mistakes.
Errors by license plate readers can lead to dangerous encounters
The expanding use of automated license plate recognition has coincided with a number of troubling incidents in which innocent motorists have been stopped, detained or confronted by armed police after technology incorrectly identified their vehicles.
In Wisconsin, for example, a woman was recently pulled over on a busy interstate after a Flock alert incorrectly associated her vehicle with a homicide. Police surrounded her with weapons drawn. According to her account to local media, it was the second time in less than two weeks that she had been stopped at gunpoint because of an apparent identification error.
The incident was not isolated. Automotive journalist Joel Feder and his wife reportedly encountered a similar situation in Minnesota while test-driving a Range Rover. Flock's system had incorrectly identified the vehicle as stolen, resulting in armed officers surrounding them.
The Institute for Justice examined 24 traffic stops associated with license plate-reader errors in a July analysis. According to the organization, police officers drew their firearms during roughly two-thirds of those encounters. The concern is not simply that the technology can produce inaccurate alerts; an incorrect alert can transform an ordinary traffic stop into a potentially life-threatening confrontation.
Traffic stops themselves carry significant risks, regardless of the reason for the stop. The American Civil Liberties Union reports that more than 800 people died during traffic-stop encounters between 2017 and 2025. It also points to racial disparities, noting that Black Americans, despite representing approximately 12 percent of the U.S. population, accounted for more than one-quarter of those deaths.
One particularly serious case occurred in Ohio earlier this year. A Black motorist was stopped after a Flock alert incorrectly associated his vehicle with a wanted vehicle. Police approached him with guns drawn, and he was subsequently attacked by a police dog, sustaining severe injuries. He was also detained in jail for several hours before authorities released him.
The scale of the system makes even a relatively small error rate significant. Flock has promoted an accuracy rate of about 93 percent, but its network reportedly processes more than 20 billion license plates each month. Even a seven-percent error rate applied to that enormous volume would translate into a very large number of inaccurate readings.
Overall accuracy figures can also obscure serious problems in particular locations. In Roseville, California, for instance, a Flock system generated 1,427 alerts involving vehicles believed to be stolen or connected to felony investigations between 2023 and 2024. More than 70 percent of those alerts were subsequently determined to be incorrect. Flock reportedly attributed the unusually high error rate to older equipment being used in a configuration requested by local authorities.