Bangladesh uses AI to police Dhaka traffic

DHAKA, Bangladesh (MNTV) — Bangladesh has launched an AI-powered traffic enforcement system in capital Dhaka, as authorities attempt to restore order to one of the world’s most congested and chaotic megacities.

The city of more than 22 million people has long been notorious for crippling traffic jams, weak enforcement and widespread disregard for road rules, with buses, motorcycles, cars and pedal rickshaws competing for limited road space in near-constant gridlock.

According to media reports, Dhaka police in April connected traffic surveillance cameras to artificial intelligence software capable of automatically detecting traffic violations, including running red lights, illegal parking and lane breaches.

Authorities say the system is already beginning to change driver behavior in the city where manual traffic control has historically dominated intersections, often requiring officers to physically direct vehicles or stretch ropes across roads before traffic lights change.

“Those who violate rules turn against us,” traffic sergeant SM Nazim Uddin told the media, describing frequent confrontations between motorists and officers during roadside enforcement operations.

“But since AI was introduced, people behind the wheel have started obeying law — and we have been spared everyday quarrels,” he added.

Motorist Hannan Rahman Jibon said he became more cautious after receiving an automated fine of 2,000 Bangladeshi taka ($16) for running a red light. Notification was sent directly to the vehicle owner through text message, even though the owner was not present at the scene.

Research has repeatedly ranked Dhaka among world’s worst cities for traffic congestion. One study by the US National Bureau of Economic Research described Dhaka as “slowest city” globally, while research conducted by World Bank and Bangladesh University of Engineering and Technology found average traffic speed in the city had fallen to just 4.8 kilometers per hour — slower than walking pace.

Police spokesman N.M. Nasiruddin said the AI system relies on existing traffic camera infrastructure, with human operators reviewing flagged violations before penalties are issued. Authorities say at least 300 vehicles have already been prosecuted under the new system, while hundreds of other offenders received warnings.

Inside police headquarters, analysts monitor live video feeds linked to AI software capable of detecting violations in real time. Officials say nearly 800 violations were recorded in a single day during recent monitoring.

However, authorities acknowledge major challenges remain. Some vehicle number plates are difficult to identify due to poor visibility or non-standard formatting, while enforcement still does not fully cover Dhaka’s sprawling urban network.

The system currently focuses mainly on motor vehicles, leaving unresolved the question of how to regulate the city’s massive fleet of pedal rickshaws, which remain central to Dhaka’s transportation system.

Hasib Mohammed Ahsan, professor at Bangladesh University of Engineering and Technology, warned that technology alone would not solve the city’s traffic crisis unless authorities maintain consistent enforcement and accountability.

“We do not follow rules, there is no consistency in our planning, and there is no accountability for failure,” Ahsan said.