This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
A Kansas couple was held at gunpoint and handcuffed by Oklahoma police after a Flock automatic license plate reader camera misidentified their specialized license plate as a stolen vehicle due to a system limitation.
Matthew and Leah Spurlock were reportedly stopped in Colwich, Kansas, and about a month later in Catoosa, Oklahoma, after Flock Safety cameras associated their KU alumni plate 7007 with a stolen GMC pickup bearing a temporary Kansas tag with the same number. Colwich police quickly resolved the first stop; in Catoosa, the couple said officers drew guns and handcuffed Matthew. Wichita police said Flock reads plate characters but not plate type.
Domain classification, causal taxonomy, severity scores, and national security assessments were LLM-classified and may contain errors.
AI systems that fail to perform reliably or effectively under varying conditions, exposing them to errors and failures that can have significant consequences, especially in critical applications or areas that require moral reasoning.
Also classified as
Police officers in Catoosa overrelied on the automated Flock camera alert to conduct a high-risk, armed stop without first performing the required manual verification of the vehicle make and model.
The Flock camera system failed to perform robustly because it lacks the capability to distinguish between different license plate backgrounds with identical alphanumeric characters, leading to false stolen-vehicle alerts.
AI system
Due to a decision or action made by an AI system
Unintentional
Due to an unexpected outcome from pursuing a goal
Post-deployment
Occurring after the AI model has been trained and deployed
The incident was caused by the Flock AI camera system failing to distinguish license plate backgrounds post-deployment, leading to an unintended high-risk police stop.
Added categories peak at 2: Minor (Psychological Harm, Direct)
An automated license plate reader failed to distinguish between unique state plate backgrounds with identical alphanumeric characters, leading police to conduct a high-risk traffic stop on innocent citizens. This incident highlights domestic civil liberties and local law enforcement verification challenges rather than a direct threat to national security.
The Flock camera system acts as a tool that performs a specific, well-defined function: taking pictures of license plates, reading the alphanumeric characters, and matching them against a database of stolen vehicles without autonomous planning or context retention.
High Risk: The system is used in law enforcement applications (automatic license plate recognition for identifying stolen vehicles), which has significant implications for public safety and fundamental civil rights.
The primary function of the Flock cameras is to capture and read license plates to detect vehicles associated with crimes or reported stolen.
The report provides clear, first-hand accounts from the victims, statements from local police departments, and technical explanations of how the Flock cameras operate and why the error occurred.
None
Technology
No Plate Background Recognition
AI system only reads alphanumeric characters, ignoring specialty plate designs.
Inability to Distinguish Plate Types
Flock cameras cannot differentiate between standard, KU, or veteran tags.
Data Inputs
Duplicate License Plate Numbers
Kansas state issues identical alphanumeric combinations across different plate types.
Incorrect Plate Type Entry
Human operators entered the incorrect plate design category into the database.
Human Factors
Overreliance on Camera Alerts
Officers acted aggressively on automated alerts without manual visual verification.
Failure to Verify Vehicle Match
Officers did not verify if the vehicle type matched the stolen vehicle record.
Process and Methods
Inadequate Alert Verification
Lack of robust secondary validation protocol before initiating high-risk stops.
Systemic Database Entry Errors
The process allows incorrect vehicle or plate metadata to be saved as active.
Management
Premature System Deployment
Deploying ALPR technology in areas with duplicate plates without safeguards.
Lack of Operator Training
Insufficient training on how to handle and verify automated plate reader alerts.
Regulatory Environment
Permissive Plate Duplication Policy
State regulations allowed duplicate numbers across dozens of specialty plates.
AI system involved
Identified by automated extraction; can differ from the alleged developers above.
