ALPR Evidence in Criminal Cases

What counts as ALPR evidence

“The camera saw the car” usually stands for several distinct records. The California State Auditor describes the core of every ALPR system the same way: software extracts the plate number from the image and stores it with the date, time and location of the scan in a searchable database. Around that core, a case can involve:

• The read record: image, plate text, time stamp, camera location, and for Flock, vehicle characteristics such as make, color and body type. See what data license plate readers collect.
• The alert: a match between the plate text and a hotlist entry, and the hotlist entry itself. See ALPR hotlists and alerts.
• Search records: who searched the stored data, when, and for what stated reason.
• What people did with it: dispatch communications, verification steps, reports, and video of any stop.

How a camera record enters a criminal case

The system works in a fixed sequence: capture, plate read, time and place, hotlist check, alert, search, and eventually deletion. Each step is explained in how Flock cameras work. In practice, the records tend to reach a case in one of three ways:

1. A live alert prompts a stop. A plate read matches a hotlist, an alert goes out, and officers stop the vehicle. The central questions are whether the read and the list were right, and whether officers verified the alert before acting.
2. An investigator searches stored data. After a crime, police search past reads to find a vehicle or trace where it traveled. Knoxville’s police policy, for example, lists “identifying an unknown suspect” and “locating travel patterns of a vehicle on a hot list” among its investigative uses. The central questions are what was searched, by whom, on what basis, and what the results actually show.
3. The records are offered to show a vehicle’s location. A read may be used to place a vehicle near a scene at a particular time. The central questions are whether the record is accurate and complete, and what links a person to that vehicle.

Where ALPR evidence can be questioned

Because the evidence is a chain of separate records, a problem at any link can matter. This table maps the common issues to the guide that covers each one.

If a problem affected how evidence was obtained, it may be raised in a motion to suppress. Whether that’s possible depends on the facts and the law where the case is brought.

What published data says about reliability

There is less independent data than you might expect. We found no independent, audited field test of Flock’s per-read accuracy. Flock reports OCR accuracy above 96 percent without a published method. Agency records measure something different, the share of alerts that were wrong, and those figures can be high. Roseville, California, reported that 71 percent of 1,427 stolen- or felony-vehicle alerts in 2023–2024 were misreads. Flock disputes that figure. The explanation, and every source with its limits, is in license plate misreads and ALPR accuracy.

Vendors and agencies agree on one point: alerts need human confirmation. Flock’s own policy says users “should confirm the computer translation prior to taking any action.”

Why timing matters

ALPR data that isn’t preserved can disappear quickly. Retention is set by the customer agency. In August 2026, Flock lowered its recommended default from 30 days to seven, and agencies with an approved period kept it. Once deleted, Flock says, data is permanently gone unless it was preserved as evidence. Which records exist in a given case, and whether they still exist, is something a defense attorney will want to establish early. Requesting ALPR records in discovery covers how that works.

Flock and ALPR: what’s vendor-specific

Flock Safety is one ALPR vendor among several. The basic mechanics on this site (capture, plate read, hotlist check, alert) apply to ALPR systems generally. Features like vehicle-characteristic search, specific retention defaults, and search-logging rules are vendor- and agency-specific. Throughout this guide we label which is which, and we cite the vendor’s own documentation for vendor claims. Flock Camera Defense is independent and not affiliated with Flock Safety.