This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
Nijeer Parks, a Black man in New Jersey, was wrongfully arrested and imprisoned for 10 days after law enforcement relied on a faulty facial recognition match to identify him as a suspect in a shoplifting and assault incident. Despite having a solid alibi and no connection to the crime scene, Parks was forced to spend his savings on legal fees to clear his name. The case highlights the risks of racial bias and inaccuracy in facial recognition technology used by law enforcement.
Woodbridge Police Department falsely arrested an innocent Black man following a misidentification by their facial recognition software, who was jailed for more than a week and paid thousands of dollar for his defense.
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 overrelied on the facial recognition match as definitive proof of identity, ignoring the lack of supporting forensic evidence and the user's alibi.
The facial recognition software exhibited unequal performance across demographic groups, specifically failing to accurately identify Black faces, which directly led to the wrongful arrest of Nijeer Parks.
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 false match produced by the facial recognition AI system during its active deployment.
Added categories peak at 2: Minor (Psychological Harm, Direct)
A domestic law enforcement incident where a faulty facial recognition match led to the wrongful arrest and 10-day imprisonment of a Black man in New Jersey. While highlighting critical civil rights and algorithmic bias concerns, the national security impact remains negligible to minor, with no threat to physical infrastructure, sovereignty, or national economic systems.
The facial recognition software acted as an assistant to law enforcement, providing suggested matches to help identify suspects, which required human verification and decision-making before taking action.
High Risk: The AI system is used in law enforcement applications, specifically facial recognition, which is explicitly classified as High Risk under the EU AI Act.
The primary function of the AI system was to perform facial recognition to match a suspect's photo against a database of images.
The reports provide consistent, detailed accounts of the incident from multiple reputable news sources, detailing the timeline, the victim's experience, and the broader context of facial recognition failures.
None
Technology
Algorithmic Demographic Bias
FRT performs poorly on Black faces, leading to false matches.
Flawed Facial Recognition Matching
The algorithm incorrectly matched Parks to a fake Tennessee license.
Biased Pretrial Risk Assessment
Risk assessment algorithm kept Parks jailed due to past history.
Data Inputs
Low-Quality License Photo
The system used a low-quality photo from a fraudulent license.
Unverified Database Matches
Matches were not validated against physical evidence like fingerprints.
Human Factors
Confirmation Bias by Detective
Detective agreed with the FRT match without objective verification.
Erroneous Witness Identification
Hertz employee wrongly confirmed the suspect matched Parks' photo.
Overreliance on AI Output
Officers treated the algorithmic match as definitive proof of identity.
Process and Methods
Sole Reliance on FRT for Warrant
Warrant issued based solely on FRT match without corroborating forensics.
Lack of Forensic Verification
Police failed to check fingerprints or DNA left at the crime scene.
Inadequate Interrogation Procedures
Interrogators ignored Parks' alibi and lack of driver's license.
Management
Failure to Enforce Lead-Only Policy
Management failed to ensure FRT was used only as an investigative clue.
Inadequate Officer Training
Officers were not trained on the limitations and errors of FRT.
Regulatory Environment
Absence of FRT Regulations
Lack of clear state policies or bans on facial recognition before 2019.
Lack of Algorithmic Oversight
No regulatory body monitored the accuracy of police facial searches.
AI system involved
Identified by automated extraction; can differ from the alleged developers above.
