In April, Dhaka police introduced an AI-powered traffic monitoring system that links roadside cameras with software capable of automatically detecting traffic violations.
In Bangladesh’s capital, traffic rules often struggle to compete with the reality on the streets. Buses squeeze through impossible gaps, motorcycles weave across lanes, rickshaws push through crowded junctions and frustrated drivers ignore signals as if they barely exist. In a city where traffic jams have become part of daily life, even crossing a road can feel like a risk.
Now, authorities in Bangladesh’s overcrowded capital are betting on artificial intelligence to restore some order to its streets — and police say the technology is already beginning to change driver behaviour.
In April, Dhaka police introduced an AI-powered traffic monitoring system that links roadside cameras with software capable of automatically detecting traffic violations, according to news agency AFP. Officials say the system identifies offences such as running red lights, illegal parking and lane violations without requiring officers to physically stop vehicles on the road.
It is one of the most serious attempts in years to modernise a traffic management system long criticised as outdated, inconsistent and heavily dependent on manual enforcement.
“Those who violate the rules turn against us,” traffic sergeant SM Nazim Uddin told AFP.
“But since AI was introduced, people behind the wheel have started obeying the law — and we have been spared the everyday quarrels.”
A CITY PARALYSED BY TRAFFIC
The Bangladeshi capital has earned a reputation for severe traffic chaos.
One study by the US National Bureau of Economic Research ranked the city as the slowest in the world. According to research by the World Bank and Bangladesh University of Engineering and Technology, average traffic speed in the capital stands at just 4.8 kilometres per hour — slower than walking pace in many parts of the city.
Despite spending years installing traffic lights and launching road management projects, authorities struggled to maintain discipline on the streets. Officers often relied on physical barriers and hand signals to manage traffic because many drivers ignored automated systems.
The AI rollout is designed to reduce that dependence on direct confrontation between police and motorists.
The new system sends automatic notifications and fines to registered vehicle owners after violations are detected through camera feeds. Drivers no longer need to be stopped on the road for many offences.
Motorist Hannan Rahman Jibon was among the first to experience the new system.
“I ran a red light and the owner of my car, who was sitting at home, received a text message saying that the vehicle had violated traffic rules,” Jibon told AFP.
The 28-year-old driver said the experience made him far more cautious behind the wheel.
POLICE SAY BEHAVIOUR IS CHANGING
Officials monitoring the programme say the technology is already showing results.
Dhaka police spokesperson N.M. Nasiruddin said the AI software uses existing surveillance camera feeds and automatically flags possible violations before human operators verify them.
“We have started getting results,” Nasiruddin told AFP.
“We have prosecuted at least 300 vehicles.”
Inside the police control room, analysts monitor dozens of live video feeds linked to the AI system throughout the day. The software rapidly scans moving traffic, identifying suspicious driving behaviour that would otherwise require large numbers of officers to track manually.
Analyst Sharmin Afroze said the system recorded nearly 800 traffic violations in a single day. For now, authorities are issuing fines mainly to serious offenders while warning others.
TECHNOLOGY ALONE MAY NOT FIX DHAKA’S ROADS
Despite early optimism, technology still faces significant obstacles.
Some vehicle number plates remain difficult for cameras to read because they are blurred, damaged or too small. Authorities are also still trying to determine how the system can regulate Dhaka’s enormous fleet of pedal rickshaws, which remain a major part of city transport.
Additional AI features, including detection of vehicles driving on footpaths, are expected to be introduced later. But it alone will not solve Dhaka’s traffic problems unless authorities consistently enforce rules over the long term.
However, for many residents, even small improvements offer hope in a city where endless traffic congestion has long shaped daily life.
With inputs from agencies
