The information displayed in the AIM should not be reported as representing the official views of the OECD or of its member countries.
A UK police watchdog report found that retrospective facial recognition AI used by police forces has led to wrongful identifications, disproportionately affecting ethnic minorities. Officers often treat AI matches as conclusive evidence, risking wrongful arrests. Poor image quality and lack of oversight further undermine reliability and procedural safeguards.[AI generated]
Why's our monitor labelling this an incident or hazard?
The facial recognition system is an AI system used by police for retrospective searches. The report documents that officers acted on AI-generated potential matches without adequate corroboration, leading to wrongful arrests and misidentifications, disproportionately impacting ethnic minorities. This directly leads to harm in the form of violations of rights and wrongful arrests, fulfilling the criteria for an AI Incident. The involvement of the AI system is clear, and the harm is realized, not just potential. Hence, the classification is AI Incident.[AI generated]