Anyang City Deploys AI Gender Recognition CCTV in Women's Restrooms
Anyang City, South Korea announced expansion of AI gender recognition CCTV systems in women's public restrooms to detect males entering and trigger alerts to police. The system uses facial recognition and analysis to determine gender, raising concerns about surveillance, misidentification, and biometric data collection in intimate settings.
What happened
Anyang City announced on April 23, 2025, that it will expand deployment of AI gender recognition CCTV systems to 28 additional women's public restrooms in 2025.
The system recognizes faces at restroom entrances and analyzes them to determine gender. If a male is detected, an immediate alarm is sent to the city's integrated control center, displaying the restroom's location and name on screen.
Anyang had already installed the system in 45 of 85 public restrooms in parks and along waterways as of late 2024, with plans to add 28 more locations and evaluate 12 others based on budget availability.
Reported impact
The city stated that restroom crimes against women have continued, with public restrooms increasingly perceived as fearful places by citizens.
Police statistics cited in reporting showed that illegal filming incidents rose to 5,323 cases by September of the previous year, averaging 19.4 cases per day, and that public restroom sexual crimes increased from 2022 to recent years.
Index analysis
The system uses facial recognition to determine gender and trigger alerts based on perceived sex, raising concerns about potential misidentification and the collection of biometric data without explicit consent in intimate settings.
The deployment involves real-time surveillance and data transmission to a centralized control system, with integration to police notification protocols.
Governance controls
Anyang City stated that the AI gender recognition system is integrated with existing illegal filming detection systems and emergency bells to enable rapid police response when suspected crimes are identified.
Classification
Sources and evidence
Sources not yet published
This record was researched and written by the Index from original reporting. Its sources are still being verified by an editor, and they are listed here once that check is complete. Until then the record carries no reviewer.