March 20, 2026
Smile, you’re paused on camera
Essex police pause facial recognition camera use after study finds racial bias
Police hit pause on face cams as comments explode: PR cover-up or safety fix
TLDR: Essex police paused facial recognition after a study found it was better at matching some groups, sparking fairness concerns. Commenters clashed: some called it a PR smokescreen or dangerous tech, others said it’s a fixable tweak, with dark jokes and confusion over what “bias” even means.
Essex police slammed the brakes on live facial recognition (LFR) after a Cambridge study found the cameras were more likely to correctly identify Black people and men on police watchlists. The Internet immediately erupted. The top vibe? Skepticism. One user framed it as a PR timeout: Essex takes the heat while other forces keep rolling; let the outrage cool, then quietly reboot. Others were baffled by the twist: bias here isn’t about false arrests, but about the system being better at catching some groups, raising fairness questions. That confusion fueled the drama and the memes.
On Team “Shut It Down,” privacy hawks waved the red flag, echoing a watchdog’s warning that “experimental” AI shouldn’t patrol our streets. They pointed to a recent horror story: a man wrongly arrested 100 miles from home after a lookalike match. On Team “It Works, Deal With It,” commenters cited thousands of arrests for serious crimes and argued this is a settings tweak, not a scandal. The thread’s dark humor went viral, with one sarcastic “solution” suggesting the system should purposely miss matches to even things out. Meanwhile, the data nerds chimed in with, “Correlation ≠ causation,” trying to cool the flames. Cool it did not. The comments are the content, and the streets are the stage.
Key Points
- •Essex Police paused live facial recognition after identifying potential accuracy and bias risks, as disclosed by the ICO.
- •A Cambridge study of 188 actors in Chelmsford found higher correct identification rates for men and for Black participants than other groups, with false identifications extremely rare.
- •The ICO urged other forces to implement mitigations; LFR is used by at least 13 forces across England and Wales, including deployments via vans.
- •A separate National Physical Laboratory study observed similar patterns but without statistical significance; the Home Office reported 1,300+ arrests from London LFR deployments (Jan 2024–Sept 2025).
- •Critics, including Big Brother Watch, said the findings support warnings about bias; a recent wrongful arrest case involved retrospective face scanning software.