AIPolicyTracker

AI incident ·

Cross-Jurisdictional Facial Recognition Misidentification by NYPD Leads to Wrongful Arrest and Four-Year Jail Time in New Jersey

1 news report Synced from source · record last edited 4 Sep 2026

In brief

An AI system built by Facial recognition system developers and Clearview AI and deployed by West New York PD, Real Time Crime Center and 3 others allegedly harmed Francisco Arteaga.

Risk domain
AI system safety, failures, and limitations Lack of capability or robustness
Occurred
Coverage
1 reportAug 2024

What happened

In 2019, facial recognition technology misidentified Francisco Arteaga as a suspect in an armed robbery in New Jersey. The incident led to nearly four years of pretrial incarceration. Despite having an alibi, Arteaga was charged based on the flawed identification. The legal battle that followed resulted in a court ruling requiring police to reveal details about the algorithms used in facial recognition. The process exposed significant gaps in transparency and accountability.

Editor's notes

See Incident 815 for a broader overview of these specific kinds of harms. Reconstructing the timeline of events: (1) November 29, 2019: An armed robbery occurs at the Buenavista Multiservices store in West New York, New Jersey. Police submit surveillance footage for facial recognition analysis. (2) December 2019: The West New York Police Department sends surveillance footage to the NYPD's Real Time Crime Center, which identifies Francisco Arteaga as a possible match using facial recognition technology. (3) 2019-2022: Arteaga spends nearly four years in pretrial detention while fighting the charges, despite having an alibi. (4) May 13, 2022: A trial judge denies Arteaga’s motion for discovery on details of the facial recognition technology used in his case. (5) June 7, 2023: A New Jersey appellate court rules that Arteaga is entitled to information on the facial recognition technology used in his case, including the algorithm, error rates, and other relevant details.

Laws that address this harm

Policy angle: Classified under AI system safety, failures, and limitations (Lack of capability or robustness) in the MIT AI Risk Repository taxonomy; 5 recorded instruments address this use case.

Matched from the record's risk domain and country to the instruments recorded here. A reviewer can correct the match in the repository (data/external/incident_overrides.yaml).

News reports (1)

Titles link to the original publisher; report text is not reproduced here.

Who was involved

Alleged harmed party
Francisco Arteaga

AI systems implicated

Facial recognition systemsAI-enabled decision support systems

Classification (MIT AI Risk Repository taxonomy)

Causal entity
AI
Intent
Unintentional
Timing
Post-deployment
Harm level
—
Sectors
—
Countries
—

Risk entries describing this failure mode

Entries from the MIT AI Risk Repository coded to subdomain 7.3.

  • Reliability issues

    "Relying on general-purpose AI products that fail to fulfil their intended function can lead to harm. For example, general- purpose AI systems can make up facts (‘hallucination’), generate erroneous computer code, or pro...

    International AI Safety Report 2025 (Bengio2025)

  • Type 2: Bigger than expected

    Harm can result from AI that was not expected to have a large impact at all, such as a lab leak, a surprisingly addictive open-source product, or an unexpected repurposing of a research prototype.

    TASRA: a Taxonomy and Analysis of Societal-Scale Risks from AI (Critch2023)

  • Type 3: Worse than expected

    AI intended to have a large societal impact can turn out harmful by mistake, such as a popular product that creates problems and partially solves them only for its users.

    TASRA: a Taxonomy and Analysis of Societal-Scale Risks from AI (Critch2023)

  • Ethics and Morality Issues

    LMs need to pay more attention to universally accepted societal values at the level of ethics and morality, including the judgement of right and wrong, and its relationship with social norms and laws.

    Towards Safer Generative Language Models: A Survey on Safety Risks, Evaluations, and Improvements (Deng2023)

  • Safe learning

    "AGIs should avoid making fatal mistakes during the learning phase. Subproblems include safe exploration and distributional shift (DeepMind, OpenAI), and continual learning (Berkeley)."

    AGI Safety Literature Review (Everitt2018 )

  • Malign belief distributions

    "Christiano (2016) argues that the universal distribution M (Hutter, 2005; Solomonoff, 1964a,b, 1978) is malign. The argument is somewhat intricate, and is based on the idea that a hypothesis about the world often includ...

    AGI Safety Literature Review (Everitt2018 )

  • Meta-cognition

    "Agents that reason about their own computational resources and logically uncertain events can encounter strange paradoxes due to Godelian limitations (Fallenstein and Soares, 2015; Soares and Fallenstein, 2014, 2017) an...

    AGI Safety Literature Review (Everitt2018 )

  • Technical and operational risks

    "To date, technical limitations and vulnerabilities are present in most generative AI models in various contexts. Consequently, malicious users find it easier to breach an AI system’s safety and ethical guardrails to e...

    Regulating under Uncertainty: Governance Options for Generative AI (G'sell2024)

Linked by editors or by text similarity in the source dataset.

Incidents in the same risk subdomain

All incidents in this subdomain

Other incidents involving West New York PD

Source record: incident #816 on the AI Incident Database · all 1 report