Hanoi's AI traffic cameras log nearly 400 violations in three days

By Viet An   December 18, 2025 | 05:38 pm PT
Hanoi's AI traffic cameras log nearly 400 violations in three days
A motorbike driver jumpes the red light on Nguyen Trai Street, Hanoi, Vietnam on Dec. 15, 2025. Photo by VnExpress/Viet An
Hanoi logged nearly 400 traffic violations in just three days after launching an AI-powered camera system, highlighting a sharp jump in detected offenses compared with traditional patrols.

The system, operated by the Hanoi Traffic Control Center since Dec. 13, oversees more than 1,800 AI cameras installed at around 195 intersections, 25 major roads and key bridges across the capital.

Capable of detecting up to 28 types of traffic violations, the network recorded 393 cases between Dec. 13 and 16, including 125 involving cars and 268 involving motorbikes, a figure authorities say is significantly higher than what was typically identified through traditional patrols.

Running red lights accounted for the vast majority of violations, with 352 cases detected. The system also flagged cars ignoring road markings, driving the wrong way on one-way streets, stopping in prohibited areas and illegally picking up or dropping off passengers, as well as several cases of motorbikes carrying more passengers than allowed.

According to the Hanoi Traffic Police Department, early assessments show the AI camera network is operating stably and effectively, with a sharp rise in detected violations compared with earlier enforcement methods. High-resolution footage clearly captures license plates, timestamps and violation behaviors, supporting automated fine issuance while ensuring accuracy and objectivity. All data is transmitted to a centralized server and stored for up to 75 days.

Beyond traffic enforcement, the AI system has also been used for public safety.

On the evening of Dec. 15, police used camera footage to trace the movements of a 78-year-old man who went missing after leaving home for exercise in Thanh Xuan Ward. By tracking his route and sharing identifying information with the public, authorities and residents were able to locate him in the early hours of Dec. 16, when a local citizen found him weak and helped return him home safely.

 
 
go to top