Stay up-to-date with our monthly email recaps of the latest auto industry news, car reviews and driving safety, technology and lifestyle tips.
The best way to stay in touch with Car Coach reports is to sign up (above) for our monthly email news report. Then subscribe on YouTube for alerts when new video stories and car reviews publish!
Stay up-to-date with our monthly email recaps of the latest auto industry news, car reviews and driving safety, technology and lifestyle tips.
The best way to stay in touch with Car Coach reports is to sign up (above) for our monthly email news report. Then subscribe on YouTube for alerts when new video stories and car reviews publish!
What if you could prove you weren’t driving a car when it committed a traffic violation and still be responsible for the ticket?
That may sound absurd, but that is essentially the issue raised by Jacob Rockwell’s experience with a red-light camera citation in Pensacola, Florida. Rockwell said he was about 25 miles away in Seminole, Alabama, when a camera photographed his vehicle going through a red light. The $158 citation was issued to him as the registered owner, even though he said someone else was driving. Rockwell challenged the citation and later raised his concerns publicly about how automated traffic enforcement assigns responsibility.
The case highlights a basic problem with automated traffic enforcement: A license plate identifies a vehicle. It does not necessarily identify the person driving it.
That distinction becomes increasingly important as automated license plate recognition, or ALPR, systems become more common.
There is an important difference between what an automated camera records and what authorities may conclude from that information.
A camera can capture a license plate. It can record the time and location of a vehicle. Some systems can also identify characteristics such as the vehicle’s make, model and color.
But none of those things, by themselves, necessarily establish who was behind the wheel.
That creates a potential chain of assumptions:
Camera sees vehicle → database identifies registered owner → authorities contact owner → owner must explain why he or she wasn’t responsible.
The technology may have accurately recorded the vehicle. The problem can arise when the information is treated as proof of who committed an offense.
And there are documented cases showing what can happen when that distinction gets lost.
Consider the case of Chrisanna Elser in Colorado.
Police accused Elser of stealing a package after Flock Safety camera images showed her vehicle traveling through the area around the time of the theft. Elser disputed the accusation and eventually gathered evidence, including vehicle and location information, to establish that she was not responsible.
The case was ultimately dropped. Reporting on the incident showed how a camera image that correctly placed a vehicle in an area could nevertheless contribute to an incorrect accusation against the person associated with that vehicle.
That’s an important distinction.
The camera didn’t necessarily get the vehicle wrong. The problem was what people concluded from the camera’s information.
That is precisely why automated license plate readers should be treated as investigative tools rather than automatic proof of guilt.
The consequences become dramatically more serious when an ALPR hit is used in a criminal investigation.
Lindsey Isaacs learned that firsthand.
In 2025, a fatal crash on Interstate 4 in Florida killed three people. Investigators were looking for a Dodge Durango. A Flock camera captured Isaacs’ black Dodge Durango near the area shortly before the crash.
Investigators subsequently focused on Isaacs. She was arrested and spent 13 days in jail while facing multiple felony charges.
The problem was that she was not the driver involved in the fatal crash.
A subsequent investigation determined that the vehicle involved was a maroon Dodge Durango, not Isaacs’ black vehicle. Prosecutors dropped the charges, and another suspect was arrested.
That case illustrates the difference between a lead and evidence.
The camera produced information that investigators considered relevant. But the existence of a camera image did not establish that Isaacs’ vehicle was the vehicle involved in the crash.
The distinction may sound obvious.
It becomes much less obvious when technology produces information quickly, appears precise and places a vehicle in a database with a timestamp and location.
A computer-generated result can look authoritative even when the underlying conclusion is wrong.
Automotive journalist Tim Esterdahl experienced another example of how an automated license plate system can create problems.
In July 2026, Esterdahl reported that a Flock system incorrectly identified the Jaguar Land Rover press vehicle he was driving as stolen. The resulting traffic stop lasted about 90 minutes while officers worked to determine what had happened. Esterdahl reported that the system had apparently connected information about a stolen manufacturer plate with other plates sharing similar characteristics.
The important point is that the responding officer was not necessarily the source of the original error.
The officer received an alert and had to investigate it.
That illustrates a larger problem with automated enforcement and surveillance systems:
An error can begin with software but become a very real problem for a human being.
The computer doesn’t get handcuffed. Instead, the driver may spend hours on the side of the road. Ultimately, it’s the person behind the wheel who has to explain why the vehicle was somewhere.
The person does.
That brings us to Flock Safety, one of the most prominent companies in the automated license plate reader industry.
On Aug. 26, 2026, U.S. Sen. Josh Hawley announced an investigation into Flock Safety’s collection, retention and dissemination of information generated by its AI-powered camera network. Hawley’s office said the company has more than 120,000 cameras across 49 states and that the network records more than 20 billion vehicle scans each month.
Those numbers are enormous.
And they raise questions that go well beyond whether a camera can identify a stolen vehicle.
What information is being collected?
How long is it retained?
Who can search it?
Can one law enforcement agency access information collected by another?
Under what circumstances can searches be conducted?
What safeguards exist when a vehicle is incorrectly identified?
And perhaps most importantly:
Who is accountable when the system is wrong?
Those questions aren’t limited to Flock Safety. They apply to the entire category of automated license plate recognition technology.
The name on the camera can change. The fundamental questions do not.
If taxpayers are told that automated license plate readers make communities safer, the public should be able to understand how these systems operate.
That means examining contracts, costs, retention policies, access rules and auditing procedures.
It also means understanding what happens after a camera generates a questionable identification.
Is the alert independently verified?
Is there a requirement to check physical evidence?
Can officers search the database using broad descriptions such as vehicle color and model?
Are searches logged and audited?
What happens when an innocent person is incorrectly identified?
Those questions matter because most vehicles captured by these systems belong to people who aren’t suspected of committing crimes.
They are people going to work.
Parents taking children to school.
People going to the grocery store.
Families traveling across town.
Workers driving home.
Their vehicles can become data points in a surveillance system without them ever knowing a camera recorded them.
None of this means law enforcement should stop using technology.
Automated license plate readers can provide investigators with leads. They can help locate vehicles and reconstruct movements during investigations. Used appropriately, that information can be valuable.
The issue is what happens after the technology produces a result.
A camera should be a tool.
A database should be a tool.
An AI-generated match should be a lead.
None should automatically become a verdict.
That distinction is particularly important as artificial intelligence becomes more deeply integrated into surveillance and law enforcement systems.
The more powerful these systems become, the greater the need for human verification.
The fundamental question raised by these cases is not whether technology is good or bad.
It is whether the government should be allowed to shift the practical burden of correcting a technological mistake onto the person who was incorrectly identified.
Jacob Rockwell’s traffic citation illustrates the issue at its most basic level. The camera captured his vehicle. Rockwell said he wasn’t driving it. The disagreement was not necessarily about whether the vehicle ran the red light. It was about whether the registered owner should automatically bear responsibility for what the camera recorded.
Chrisanna Elser’s experience demonstrates how an accurate vehicle identification can contribute to an inaccurate accusation.
Tim Esterdahl’s traffic stop demonstrates how an incorrect database match can escalate into a significant real-world encounter.
And Lindsey Isaacs’ case demonstrates just how severe the consequences can become when investigators treat a vehicle identification as sufficient to identify a criminal suspect.
These aren’t arguments against technology.
They’re arguments for accountability.
A camera doesn’t go to court.
An algorithm doesn’t lose a job.
A database doesn’t spend 13 days in jail.
A software system doesn’t have to explain itself to a police officer on the side of the road.
A person does.
That’s why the expansion of automated license plate recognition deserves careful scrutiny. The technology may provide valuable information, but information isn’t the same thing as proof.
A license plate identifies a vehicle.
It does not necessarily identify the driver.
A camera can identify where a vehicle was.
It does not automatically establish why it was there.
And an AI system can generate a match.
That doesn’t mean the match is the truth.
The standard should be simple: Show me the evidence. Explain how the identification was verified and who had access to the information. Tell me how long that information was retained. And if the system gets it wrong, tell me who is responsible for fixing the damage.
Because when a machine makes the accusation, the most important question isn’t whether the machine can be wrong.
We already know it can.
The question is who pays the price when it is?
You can support my work by buying me a cup of coffee:
Looking for more automotive news?
Listen to The Drive Car Show:
Lauren Fix is an automotive expert and journalist covering industry trends, policy changes and their impact on drivers nationwide. Follow her on X at @LaurenFix for the latest car news and insights.
