When AI Starts Writing Tickets: How Cities Are Using Artificial Intelligence to Enforce Traffic Laws
Traffic cameras have been watching drivers for decades, but historically they’ve been little more than automated tripwires. Traditional red-light or speed cameras captured a photo when a car crossed a sensor or radar threshold. Human review and manual processing still played a large role.
Today, that system is evolving rapidly. Cities around the world are deploying AI-powered traffic enforcement systems that do much more than capture a photo. These systems use machine learning and computer vision to analyze traffic behavior in real time—detecting multiple violations simultaneously and automatically generating citations.
From Los Angeles to Singapore, and from Philadelphia to Athens, artificial intelligence is becoming the newest traffic cop on the street.
This shift raises important questions:
-
Which cities are deploying AI traffic enforcement?
-
What violations can AI detect?
-
Does it actually improve safety?
-
And what are the privacy concerns?
Below is a deep dive into how AI is transforming traffic enforcement.
The Evolution of Traffic Cameras
The first automated traffic enforcement systems appeared in the 1990s and early 2000s. Early systems were limited to detecting one type of violation—usually speeding or red-light running.
They worked with simple triggers:
-
Induction loops in the pavement
-
Radar sensors
-
Motion detectors
If a car crossed the sensor while speeding or after the light turned red, a photo was taken.
AI cameras are fundamentally different.
Instead of relying solely on sensors, they use computer vision models trained to interpret video footage, allowing them to detect patterns and behaviors in traffic.
Modern AI systems can detect:
-
Speeding
-
Red-light violations
-
Illegal turns
-
Drivers using phones
-
Seatbelt violations
-
Bus lane violations
-
Illegal parking
-
Vehicles driving the wrong direction
Some systems even predict potential crashes by analyzing traffic patterns.
According to industry research, AI traffic enforcement is becoming central to smart city transportation systems built around real-time data and predictive analytics.
Cities Already Using AI Traffic Enforcement
AI traffic enforcement is not a theoretical future technology—it is already being deployed in cities across the United States and globally.
Below are some of the most notable examples.
Los Angeles, California
Los Angeles is one of the largest U.S. cities experimenting with AI-assisted enforcement.
In 2025, the Los Angeles Department of Transportation began using AI cameras mounted on Metro buses to detect vehicles illegally parked in bus lanes.
The system scans video footage as buses drive through the city and automatically flags violations.
In just one month of operation:
-
Nearly 10,000 parking citations were issued automatically.
The city is also preparing to expand automated enforcement under California’s AB 645 law, which allows AI-based speed camera programs in selected cities such as Los Angeles, San Francisco, and San Jose.
San Francisco, California
San Francisco already operates a large network of red-light enforcement cameras and is gradually integrating AI capabilities.
The San Francisco Municipal Transportation Agency uses automated camera systems to detect illegal turns and red-light violations and has begun expanding these systems citywide.
The next generation of these systems will incorporate machine learning algorithms capable of identifying additional behaviors like:
-
Blocking intersections
-
Bus lane violations
-
Illegal right turns
Philadelphia, Pennsylvania
Philadelphia is experimenting with AI enforcement in school safety programs.
AI cameras have been installed on school buses and public transit vehicles to detect drivers who fail to stop when a school bus extends its stop arm.
These systems automatically record violations and issue fines to offending drivers.
The program aims to improve safety around schools where traditional enforcement is difficult due to staffing limitations.
Prince George’s County, Maryland
After two children were killed in a traffic incident near a school, Prince George’s County deployed AI-powered stop-sign enforcement cameras.
These systems detect drivers who roll through stop signs near school zones and automatically record the violation.
The initiative became the first automated stop-sign enforcement system of its kind in the state.
Santee, California
The small city of Santee near San Diego is testing a system called NoTraffic, an AI platform that monitors intersections.
The system uses cameras and radar to detect:
-
Vehicles
-
Pedestrians
-
Cyclists
It can adjust traffic signals in real time and identify dangerous behaviors like red-light running.
AI Traffic Enforcement Around the World
AI enforcement technology is spreading globally, often more quickly outside the United States.
Singapore
Singapore is one of the world leaders in smart traffic systems.
AI traffic cameras are integrated into the country’s Smart Nation initiative, combining predictive analytics, high-resolution cameras, and connected traffic signals.
The system can identify violations while also adjusting traffic signals dynamically to reduce congestion and prevent crashes.
Greece
A pilot program in Athens demonstrated just how powerful AI enforcement can be.
During testing:
-
One AI camera issued over 1,000 traffic fines in just four days.
The system detected multiple violations including:
-
Phone use while driving
-
Seatbelt violations
-
Speeding
Across eight cameras in the pilot program, more than 2,500 violations were detected in the same period.
Vietnam
Vietnamese cities are also deploying AI traffic surveillance systems.
For example, Ho Chi Minh City installed a network of AI cameras capable of detecting:
-
Red-light violations
-
Wrong-way driving
-
Lane encroachment
-
Vehicles driving on sidewalks
The system can monitor multiple violations simultaneously across the city.
What Makes AI Traffic Cameras Different?
AI traffic enforcement systems rely on deep learning computer vision models.
These systems process video frames and identify objects such as:
-
Vehicles
-
Pedestrians
-
Traffic lights
-
License plates
-
Road markings
Once the system identifies objects, it tracks them across frames to analyze behavior.
For example, an AI camera can determine:
-
Whether a vehicle stopped at a stop sign
-
If a driver crossed a lane marking illegally
-
Whether a driver is holding a phone
These systems often combine several technologies:
-
Automatic License Plate Recognition (ALPR)
-
Radar or LiDAR sensors
-
Computer vision algorithms
-
Cloud processing
Together, these tools create a powerful automated enforcement platform.
Why Cities Are Turning to AI
There are several reasons cities are rapidly adopting AI enforcement.
1. Enforcement Scale
Police departments simply cannot monitor every intersection or traffic violation.
AI cameras operate 24 hours a day and can monitor multiple lanes simultaneously.
One AI camera may observe thousands of vehicles per hour.
2. Cost Savings
Hiring police officers for traffic enforcement is expensive.
AI systems can monitor roads with far fewer personnel.
Cities often justify the investment by citing:
-
reduced staffing costs
-
increased ticket revenue
-
improved traffic safety
3. Improved Safety
Research shows automated speed cameras can significantly reduce accidents.
A study of New York City speed cameras found:
-
14% reduction in crashes after deployment.
By increasing the perceived risk of being caught, automated enforcement can change driver behavior.
The Privacy Debate
Despite the safety benefits, AI traffic enforcement is controversial.
Critics raise concerns about:
Mass Surveillance
AI cameras can record large amounts of data about drivers and vehicles.
Some systems capture:
-
license plates
-
vehicle color
-
vehicle model
-
travel patterns
Algorithmic Bias
If AI systems are trained on biased datasets, enforcement could disproportionately affect certain communities.
Lack of Oversight
Automated systems may issue tickets incorrectly, and appealing them can be difficult.
These concerns have sparked lawsuits and legislative debates in several cities.
The Future of AI Traffic Enforcement
The next generation of AI enforcement may be even more powerful.
Emerging systems are capable of detecting behaviors like:
-
distracted driving
-
aggressive driving patterns
-
tailgating
-
failure to yield to pedestrians
Some cities are experimenting with predictive enforcement, where AI identifies high-risk areas before crashes occur.
As smart city infrastructure expands, traffic enforcement may become fully integrated into urban data networks.
In the future, traffic lights, cameras, and connected vehicles may all communicate with each other to enforce traffic laws automatically.
The Road Ahead
AI traffic enforcement represents one of the biggest shifts in road policing in decades.
Supporters argue that AI can dramatically improve road safety while allowing police to focus on more serious crimes.
Critics warn that automated surveillance could lead to over-policing, privacy violations, and ticketing systems that prioritize revenue over safety.
What is clear is that the technology is already spreading quickly.
Cities across the United States—and around the world—are experimenting with AI cameras capable of issuing tickets automatically, detecting complex traffic violations, and analyzing driver behavior in real time.
For drivers, that means one thing:
the next traffic ticket may not come from a police officer.
It might come from an algorithm.
