{
  "id": 4071423,
  "title": "How AI‑powered policing is endangering the public's trust in police",
  "url": "https://urgent.news/2026/08/29/how-ai-powered-policing-is-endangering-the-publics-trust-in-police",
  "topic": "ai",
  "section": "AI",
  "published": "2026-08-29T00:40:02.000Z",
  "source": {
    "name": "Phys.org",
    "slug": "phys-org",
    "url": "https://phys.org/news/2026-08-aipowered-policing-endangering-police.html"
  },
  "original_language": "en",
  "account": "Since the debut of ChatGPT in 2022, artificial intelligence (AI) has become a prominent force in society. Its adoption has swiftly expanded into various fields, including policing, often lacking proper oversight such as clear policies or laws. The unrestricted application of AI in law enforcement raises significant concerns, as it threatens the credibility of police work and may undermine the fairness of the justice system. This is because the public cannot be confident that evidence presented by the police is not tainted by AI. Procedural justice, or the perception of fairness, is crucial for police legitimacy and is upheld through impartial law enforcement and transparent communication with the public and courts. Research has shown that AI programming is prone to bias and misinformation, sometimes even more so than reality. While discussions on AI bias and its implications persist, it is evident that the selective use of AI by police is inherently biased and must be halted immediately. In the United Kingdom, a senior detective improperly utilized AI to fabricate evidential material in several cases. Similarly, in the United States and Canada, officials have publicly shared AI-generated images of drug seizures. The integration of AI into facial recognition and surveillance systems has also sparked apprehensions regarding misidentification, as AI tends to make more errors when identifying women and individuals with darker skin tones. Regrettably, police in North Dakota recently misidentified a Tennessee grandmother due to AI facial recognition, resulting in her wrongful imprisonment for over five months. In Maryland, an officer drew a gun on a teenager after an AI-driven surveillance system falsely identified a bag of chips as a potential firearm. These incidents, among others, underscore the perils of AI use in law enforcement. Despite the recognized risks, law enforcement agencies have, inexplicably, proceeded with this inherently flawed technology. The RCMP's body camera program in Canada recently completed a pilot project utilizing AI to generate police reports. This technology, a modified version of ChatGPT, has raised significant concerns about the reliability of police reports, which are often crucial pieces of evidence in trials occurring months or years after the event. Recently, an AI-generated police report in Utah declared that an officer had transformed into a frog during an encounter due to background music from Disney's \"The Princess and the Frog\" on the officer's body camera. Only then did a police administrator acknowledge the need to rectify AI-generated reports. However, what about the countless mundane details of police reports that could potentially be fabricated by AI and go unnoticed? One of the many unresolved concerns is the accuracy with which officers review AI-generated reports, particularly the threat of wrongful convictions or dismissals based on false AI police reports. Additionally, there are numerous inherent privacy risks associated with AI technologies. The public lacks control over interactions with law enforcement when their images, voices, and personal information are recorded via body cameras and fed into AI systems without their consent. It remains unclear whether data collected by police will be utilized to train and develop subsequent iterations of AI for law enforcement purposes. The lack of consistent models for AI adoption and implementation across Canadian police services leaves the public with limited knowledge about how these technologies and their data will be utilized. Some police services have even declined to discuss their AI policies when confronted by media outlets. Police service boards have tacitly approved the adoption of AI platforms, despite the infancy of existing governance models for these technologies. The entwining of public police forces and for-profit software companies signifies an emerging economic relationship that is fundamentally altering how police engage with the public. Current oversight models for police use of AI are insufficient to address the substantial legal, ethical, and societal risks associated with these endeavors. What measures can be taken to address the apparent inevitability of this situation? The evidence compellingly suggests that the use of AI technologies by Canadian police should be prohibited. It is unjust to equate the utilization of AI in the public and private sectors with its application in the justice system, as no other institution holds the authority to remove the freedoms of individuals accused of crimes. Law enforcement must therefore be held to the most stringent standards. The known risks associated with AI use in law enforcement are manifold. Inaccurate justice outcomes often stem from human errors, police misconduct, and systemic biases, among other factors that are challenging or impossible to effectively manage. AI should not be among these risks.",
  "summary": "Since the release of ChatGPT in 2022, artificial intelligence (AI) has dominated the cultural landscape. Uptake across various sectors, including law enforcement, quickly followed, usually with minimal foresight and little to no guardrails, like effective policies or laws, to govern AI use.",
  "key_points": [
    "AI-powered policing threatens public trust in police.",
    "AI bias and misinformation compromise procedural justice.",
    "Misidentification and wrongful imprisonment result from AI errors."
  ],
  "editors_take": "The unchecked adoption of AI in policing erodes public trust, threatens police legitimacy, and undermines the fairness of the justice system, ultimately putting citizens' freedoms and rights at risk.",
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}