This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
A Georgia man was wrongfully arrested and jailed for six days after a faulty facial recognition match by Clearview AI led police to misidentify him as a suspect in a Louisiana theft case. The arrest warrant failed to disclose the use of the technology, and no corroborating evidence was gathered. The case resulted in a $200,000 settlement in a federal civil rights lawsuit.
In November 2022, Randal "Quran" Reid was arrested in Georgia based on warrants from Louisiana that reportedly stemmed from a purportedly faulty facial recognition match using Clearview AI. Despite reportedly having never visited Louisiana, Reid was jailed for six days before authorities withdrew the warrants. No corroborating evidence had been gathered, and the technology's use was omitted from legal documents. In May 2025, Jefferson Parish settled with Reid for $200,000 in a federal civil rights lawsuit.
Domain classification, causal taxonomy, severity scores, and national security assessments were LLM-classified and may contain errors.
Accuracy and effectiveness of AI decisions and actions are dependent on group membership, where decisions in AI system design and biased training data lead to unequal outcomes, reduced benefits, increased effort, and alienation of users.
Also classified as
Law enforcement officers over-relied on the AI match, treating it as a definitive identification and failing to perform basic investigative verification before arresting the victim.
The facial recognition technology exhibited unequal performance, as research shows it has significantly higher false-match rates for Black individuals like the victim compared to white individuals.
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 initiated by a false positive match from the deployed Clearview AI system, which was an unexpected and erroneous outcome of its search function.
Added categories peak at 2: Minor (Psychological Harm, Direct)
A domestic law enforcement incident in the United States where a faulty Clearview AI facial recognition match led to the wrongful arrest and six-day detention of a Georgia citizen. The incident underscores minor national security concerns related to civil liberties and algorithmic bias in domestic surveillance, but has no broader implications for national sovereignty, critical infrastructure, or defense.
The AI system acted as an assistant by generating identification leads and suggestions for law enforcement officers, who were required to provide oversight and verify the results before taking action.
High Risk: The report describes a law enforcement application of facial recognition, which is explicitly classified as high risk under the EU AI Act due to its significant implications for fundamental rights.
The primary function of Clearview AI is to perform facial recognition by matching surveillance images against a database of scraped web photos to identify individuals.
The reports provide consistent, detailed accounts of the wrongful arrest of Quran Reid, the role of Clearview AI, the legal settlement, and the systemic issues surrounding facial recognition use by the Jefferson Parish Sheriff's Office. The facts are well-documented across multiple reputable sources.
One could argue the primary failure was human negligence (police failing to verify) rather than the AI itself, but the AI's false match was the critical trigger.
Technology
Algorithmic Racial Bias
Facial recognition has higher false match rates for Black individuals.
Inherent Software Limitations
Clearview AI yielded a false positive match on surveillance video.
Data Inputs
Low-Quality Surveillance Imagery
The match was based on low-quality security footage from a store.
Scraped Web Data
System relies on billions of publicly scraped photos from the web.
Human Factors
Lack of Investigator Verification
Detectives failed to verify suspect location or physical differences.
Automation Bias
Police trusted the software match without critical evaluation.
Blind Trust by Second Agency
Baton Rouge police trusted JPSO's identification without checking.
Process and Methods
Omission of AI Use in Affidavit
Affidavit cited a 'credible source' instead of disclosing AI use.
E-Warrant Rubber-Stamping
CloudGavel app allowed rapid warrant approval with low scrutiny.
Treating Lead as Sole Basis
Using facial recognition as the sole basis for arrest, not a lead.
Management
Lack of Policies and Training
JPSO failed to establish formal policies or training for AI tool use.
Regulatory Environment
Fragmented Regulatory Landscape
Lack of consistent laws or limits on police use of facial recognition.
AI system involved
Identified by automated extraction; can differ from the alleged developers above.
