This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
A York County man was held at gunpoint and detained by police after a Flock Safety license plate reader camera incorrectly read his license plate and falsely flagged his vehicle as stolen.
Steven Melvin, a 42-year-old York County, South Carolina, motorist was reportedly detained at gunpoint for more than five minutes after a Flock Safety automated license-plate reader misidentified the first letter of his plate and matched his BMW to a similar stolen vehicle. The sheriff’s office later said an illegal tinted plate cover caused the malfunction and found the deputy’s actions “reasonable”; Flock said alerts should be independently verified before enforcement.
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
The deputy immediately drew his weapon and detained the driver based solely on the automated alert, demonstrating unsafe overreliance on the technology without independent verification.
The primary failure lies within the AI system's lack of robustness in accurately reading license plates under real-world conditions, such as when a plate cover is present, leading to a critical false positive.
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 a post-deployment malfunction of the Flock AI camera system, which unintentionally misread a license plate character and generated a false positive alert.
Added categories peak at 2: Minor (Psychological Harm, Direct)
A local law enforcement officer held a driver at gunpoint due to a false positive from a Flock Safety license plate reader. While raising civil liberties and operational safety concerns regarding police overreliance on automated alerts, the incident remains a localized policing issue with negligible national security implications.
The Flock camera operates as a tool that performs a specific, well-defined function: scanning license plates, comparing them to a database, and outputting alerts based on direct visual inputs without independent planning or context retention.
High Risk: The system is used in law enforcement applications, specifically for automated surveillance and license plate recognition, which has significant implications for public safety and fundamental civil rights.
The primary function of the Flock camera system is automatic license plate recognition, which is used to detect threats such as stolen vehicles by matching plates against law enforcement hotlists.
The report provides clear, factual details of the specific traffic stop, including names, dates, locations, and the technical reason for the system's failure. It also provides broader context on similar failures nationwide.
None. The incident is a clear-cut case of an algorithmic false positive leading to a wrongful police detention.
Technology
Character Misidentification
The Flock camera incorrectly read the first letter of the license plate.
Algorithmic Fragility
The OCR algorithm failed to handle minor visual distortions on the plate.
Frequent Misreads
Systemic issues exist where 1 in 10 scans misread the state identifier.
Data Inputs
Obscured License Plate
A tinted license plate cover physically degraded the visual input to the AI.
Hotlist Inaccuracies
Stolen vehicle databases and hotlists sometimes contain errors or old data.
Human Factors
Lack of Independent Verification
The deputy failed to manually verify the plate before drawing his weapon.
Automation Bias
Law enforcement over-relied on the automated alert as absolute truth.
Aggressive Escalation
The officer immediately held the driver at gunpoint based solely on an alert.
Process and Methods
Flawed SOPs
Standard operating procedures did not mandate verification before high-risk stops.
Inadequate Training
Officers were not sufficiently trained to treat AI alerts as investigative leads.
Management
Defensive Organizational Culture
The sheriff's office cleared the officer of wrongdoing despite clear errors.
Inadequate Risk Assessment
Management failed to assess risks of wrongful detentions from false positives.
Regulatory Environment
Lack of AI Standards
No federal standards exist to regulate the accuracy of police AI surveillance.
Inadequate Oversight
There is minimal regulatory oversight on the deployment of automated ALPR systems.
AI system involved
Identified by automated extraction; can differ from the alleged developers above.
