The Signal Revolution
Seoul's traffic lights are no longer dumb. Since 2024, the city has deployed an AI-powered traffic management system across 12,000 intersections — and the results are remarkable. Average commute times have dropped 22%. Emergency vehicle response times are down 31%. And carbon emissions from idling vehicles have fallen by an estimated 14%.
The system, developed by Korean AI firm Mobileye in partnership with the Seoul Metropolitan Government, uses real-time data from cameras, sensors, and connected vehicles to optimise signal timing on a second-by-second basis.
Beyond the Lights
But the real innovation isn't the AI — it's the data infrastructure underneath it. Seoul spent five years building a unified urban data platform that connects traffic signals, public transit, ride-hailing services, and parking systems into a single decision-making layer.
"Most smart city projects fail because the data is siloed," says Professor Kim Ji-hyun of KAIST. "Seoul succeeded because they built the data layer first and the applications second."
Who's Next
Singapore, Shenzhen, and Dubai are all piloting similar systems. European cities — constrained by stricter data privacy regulations — are moving more cautiously. The question for Western cities is whether they can achieve similar results within their regulatory frameworks.