In May 2026, Harris County commissioners voted to renew a contract for 480 license plate cameras at a cost of just under $869,000, running through June 2027. That is the largest such network in Texas. At the same meeting, residents lined up to object.
Elsewhere in the state, the opposite happened. According to reporting by the Texas Tribune on August 24, 2026, Guadalupe County and Hood County ended their contracts. Bandera, Lago Vista, and Magnolia dropped theirs. Austin and San Marcos moved on. One Texas county pulled every camera it had after two of them were vandalized. In Conroe, residents are pushing for a November 2026 ballot measure. League City's mayor has floated the same idea.
Nationally, one count in June 2026 put the total at 53 cities and towns across 20 states that had canceled or rejected contracts with Flock Safety, the largest company in this market.
And the cameras keep going up.
If you manage a commercial property, an industrial site, or a community in the Houston area, that mix looks confusing. It should not be. Most of the public argument is aimed at the wrong thing. The controversy is not really about cameras. It is about the shared data pool behind them, and about who is allowed to search it. Those are two separate decisions, and you can make one without making the other.
A camera that reads license plates is not very powerful on its own. It tells you a gray pickup came through your lot at 2:14 a.m. That is slightly better than nothing.
The power comes from scale. When thousands of cameras across many cities feed one shared pool, a single plate becomes a trail. A vehicle that does something at your building and drives away can still be found, because it drove past dozens of other cameras in other towns. A case that would sit open for years gets a lead in minutes.
That is the actual product. Not the camera. The pool.
It also works. In Genetec's survey of law enforcement, roughly 60% of crimes involve a vehicle. When a car is the only witness, following the car is often the entire case. The Harris County Sheriff's Office pointed to recent results at the May 2026 renewal hearing, including a recovered stolen vehicle and two kidnapped children who were located.
So when an official sits through an angry public meeting and votes yes anyway, that is the math behind it. You have to grasp that part to talk honestly about the rest.
Texas is not moving in one direction. That is why the local news feels mixed.
Contracts kept or expanded, per the Texas Tribune's August 24, 2026 reporting:
Contracts ended: Guadalupe County, Hood County, Bandera, Lago Vista, Magnolia, Austin, and San Marcos. One county removed all cameras after two vandalism incidents.
Under active challenge: Conroe, where residents want a November 2026 ballot measure, and League City, where the mayor has raised the same option.
The pattern is not liberal versus conservative or urban versus rural. It tracks fairly closely with one thing: whether the buyer could explain, in public, who was able to search the data.
This is the part of the story that gets missed, and it matters most to the people reading this.
The Texas Tribune found that removing these cameras is complicated largely because most of them are not owned by government at all. They belong to municipal utility districts, homeowners associations, neighborhood associations, retailers, and private property owners. A county cannot remove a camera it does not own.
So in the Houston area, private owners run a large share of these cameras. They are not bystanders to a government program. East Aldine, per the same reporting, budgeted close to $1 million over five years.
If you sit on an HOA board, manage a retail center, or run a MUD, you may already own one of these systems, or you may be asked to approve one this budget cycle. The duties that come with that do not shrink because the buyer is private. Neither does the risk to your name. When residents get angry about a camera at the entrance to their neighborhood, they take it to the board, not to the county.
The feature that makes a shared pool valuable is the same feature that makes it risky. A pool solves your case because outsiders can search it. That same access lets outsiders search it for reasons the local buyer never approved. There is no setting that separates the two.
The public record on this is specific:
The ACLU has laid out a detailed case that Flock Safety gave city councils a false picture of how this access worked, more than once.
Read the anger carefully, though. It is mostly not that people were recorded. It is that the terms changed after they agreed. A council approved a local crime tool. What it received was a national search system, turned on by a default setting nobody read aloud in the meeting.
There is also a contract term worth knowing before you sign anything. The Texas Tribune reported that Flock's terms allow the company to retain data rights after a contract ends. A vendor keeping access to your records after you fire them is a term you can negotiate, or refuse, but only if you read for it.
No product choice solves this, and any vendor who claims otherwise is overselling you.
People often believe they are approving a system that catches stolen cars. But you cannot search one suspect vehicle's history without first recording every vehicle's history. Nearly all of what any plate reader stores is ordinary people driving to work, dropping kids at school, and coming home. That is not misuse. That is the mechanism.
There is a second limit. A board signs the contract. A manager picks the mounting location. Everyone who drives past is recorded and none of them voted. A better platform does not close that gap.
If either of those is a dealbreaker for your property or your community, that is a fair answer. Better to reach it now than after the install.
The shared national pool is a business model, not a requirement of plate reading. Other systems are built so the data stays with the organization that collected it.
Genetec, one of the video and plate recognition platforms DSC works with, designs its system to serve a single organization. There is no automatic national pool. Sharing exists, but the customer switches it on, decides who receives access, and can switch it off. Genetec does not sell the data, does not run a shared database across its customers, and does not offer facial recognition.
Stanford provides the clearest recent example. In late July 2026, Stanford ended its Flock contract and began installing Genetec cameras, with the changeover scheduled for early September 2026. Stanford's stated reasoning is worth reading twice: its data had never actually traveled through the national pool, but the ability was there, and the school decided not to carry that risk. Under the new setup, images sit in the cloud for under an hour, then move to Stanford's own server for 30 days.
That is what owning your data looks like when it is more than a slogan. A short retention window, storage the customer controls, and no permanent door built into the side of the system.
For a private property in Houston, that distinction matters more than it does for a police department. Your plate data is a record of your tenants, your employees, your customers, and their guests. You should be able to name every person who can search it. The list should be short.
Picking the right platform cuts the data-sharing risk. It does not settle the trust question.
In August 2026, Montreal police installed a Genetec plate camera on a city boulevard. CBC Radio-Canada reported on it. The camera came down within hours. The complaint had nothing to do with data sharing or with the vendor. It was that the public had no idea the camera was there, and officials were not straight about it when asked.
Different platform. Same outcome. The lesson for any property owner is that policy and plain notice are not paperwork you finish after the install. They decide whether the system survives its first news story or its first board meeting.
Ask these in writing and keep the answers. They apply to every brand, including the ones DSC sells.
If a vendor cannot answer these in plain language, that is useful information on its own.
The technology is the easier half. This is the half that decides how the project is remembered. You can finish every item on it before the first camera goes up.
Nearly every bad license plate reader story of the past two years traces back to a missing item on this list rather than to a bad camera.
DSC designs, installs, and services video surveillance, access control, intrusion detection, and fire alarm and life safety systems for commercial, industrial, institutional, and multi-tenant properties in Houston and across Texas. We work with several video and plate reading platforms, including Genetec, and we are an authorized Genetec partner. So we have a preference, and we would rather say so plainly than pretend we do not.
The questions above are the ones we would ask for you, whatever brand you pick. If you are weighing a plate reader for a property, a community, or a campus, we are glad to work through them with you, and sometimes the honest recommendation is to wait a year.
Call (713) 464-8407 or email info@thinkdsc.com.