This March, New York City Mayor Eric Adams took a bold step toward street safety, announcing plans to quadruple the city's red-light cameras from 150 to 600 locations. This expansion, part of the nation's largest automated enforcement network, exemplifies a fundamental shift in how cities approach traffic safety, embracing AI-powered technologies to create smarter, safer streets.
As the selected vendor managing this expanded system, Verra Mobility brings not just camera hardware but sophisticated AI-powered capabilities to identify everything from red-light runners to vehicles with deliberately obscured license plates.
AI AND MACHINE LEARNING AS STRATEGIC TOOLS, NOT STANDALONE SOLUTIONS
Automated traffic enforcement has been around for decades, and people recognize the safety benefits associated with these technologies. Those technologies have received some assistance in the past decade to improve performance — AI. When thinking of AI, we need to remind ourselves that it is not a standalone solution; AI must be incorporated thoughtfully and deliberately in a way that improves efficiency without creating new challenges. The integration of AI technologies allows us to build upon our existing infrastructure with a high degree of confidence in the decisions and actions being made.
By enhancing our current platforms with AI capabilities, we create opportunities to address the root causes of traffic incidents, rather than simply responding to incidents after they occur. Instead of implementing technology for its own sake, we focus on identifying where AI can help law enforcement make streets safer, eliminate fatalities and correct dangerous driving activity.
CURRENT AI TECHNOLOGIES AT WORK
Today's AI implementations in traffic safety rely on several interconnected technologies. Computer vision systems form the backbone of our detection capabilities, using specialized cameras with high-resolution sensors that can capture clear images in various lighting and weather conditions. These images feed into deep-learning neural networks trained to recognize vehicles, license plates, traffic behaviors and potential violations.
In the future, the processing will happen through edge computing systems installed directly at intersections or on enforcement equipment, which will allow for real-time analysis without requiring constant connectivity to central servers. For data transmission, secure 5G and LTE connections would then create a mesh network of enforcement points that share information with central traffic management systems.
REAL-WORLD RESULTS CHANGING DRIVER BEHAVIOR
Recent implementations demonstrate the practical impact of these technologies. In Fairfax County, Va., school buses equipped with stop-arm cameras are helping protect children by capturing vehicles that illegally pass stopped buses. The system activates when the stop arm extends, recording video that is reviewed by law enforcement before citations are issued.
According to local officials, even though officers approve verified violations, the technology dramatically improves compliance and helps streamline the process. At Verra Mobility, their combined school bus safety programs report that in a given school year, 98 percent of drivers who receive a citation do not get a second — reinforcing how these systems change driver behavior over time.
EXPANDING CAPABILITIES THROUGH AI INTEGRATION
With recent advances in AI technology, we're exploring additional use cases for urban safety initiatives. One promising application relates to emergency response: identifying traffic accidents in intersections and alerting emergency officials to reduce response time. Eventually, the infrastructure could warn other drivers about an accident and route all traffic away from the impacted intersection.
Another application involves addressing challenges like absconded license plates. With the growth of automated freeway and bridge tolling, some drivers attempt to evade enforcement by covering or altering their license plates with methods ranging from spray paint to plastic covers. This is also an issue when it comes to red-light running and speeding where cities have had issues holding all vehicles equally accountable to the law.
States like California have passed new laws like AB 2111 to address tech-enabled obscuring technology and make it illegal to sell license-plate obscuring devices. With AI, even if a license plate is covered, we can potentially identify violating vehicles using distinctive features of the vehicle itself. This helps communities hold the most egregious offenders accountable.
DEEPER INSIGHTS THROUGH CONNECTED DATA
By leveraging AI to analyze traffic enforcement data, we can address root causes rather than simply responding as traffic incidents happen. Using a network of different data-collecting assets like red-light and speed cameras, we will soon be able to dive deeper into the "what," "how" and "why" of traffic patterns and situations that cause unsafe driving behavior.
For example, we know that rubbernecking on highways can cause additional crashes. With more AI-enhanced data points, we can potentially reroute traffic as an incident occurs to prevent rubbernecking, reducing secondary crashes and minimizing the initial accident's impact on traffic flow.
This comprehensive approach extends to infrastructure monitoring as well. AI-enabled cameras can identify developing issues with roadways — expanding potholes, degrading lane markings or other maintenance needs — before they become safety hazards. By identifying these issues early, cities can allocate resources more efficiently and prevent accidents caused by poor road conditions.
INTEGRATING WITH EXISTING SYSTEMS
One key technical challenge in implementing these solutions is integration with existing government systems. Many jurisdictions operate with legacy infrastructure that wasn't designed for AI integration. Modern implementations use standardized APIs and middleware solutions that bridge these systems, allowing for incremental adoption without requiring complete infrastructure replacement.
Integration with court management systems, DMV databases and citation processing workflows requires careful planning and coordination. Ultimately, secure cloud-based systems are generally built from the start to integrate with different systems and APIs, so moving to a cloud system can help facilitate connections across an agency's different systems.
Cities that have already invested in camera networks are well-positioned to leverage AI capabilities. In New York City, the expansion to over 2,400 cameras creates an unparalleled opportunity to gather data and implement smart safety initiatives, providing a foundation for advanced safety applications.
PRIVACY, SECURITY AND HUMAN OVERSIGHT
As with any technology that processes potentially sensitive information, robust security measures are essential. Current implementations employ end-to-end encryption for data transmission, role-based access controls and comprehensive audit logging. Privacy-preserving techniques like automated facial redaction and selective data retention policies help balance enforcement needs with privacy concerns.
The application of AI to automated enforcement doesn't replace human judgment — it enhances it. When AI supports initial violation detection, it increases processing efficiency while maintaining the critical human review component. This allows enforcement personnel to focus their attention on complex cases that require experienced judgment.
A HOLISTIC APPROACH TO SAFER, CLEANER COMMUNITIES
While we embrace AI and its advancements, achieving truly safer streets requires a holistic approach combining policy, technology and public education. AI doesn't eliminate jobs but rather enhances how we operate in pursuit of our goals.
As we derive more data from AI-enhanced systems, we can better analyze how streets, cities and citizens interact — not just at specific points but from point to point. With a comprehensive network of data-collecting assets, we're able to derive deeper insights that inform better decision-making.
The goal remains constant: safer streets and healthier communities. By utilizing smarter technology, cities can have a more adaptive approach to enable more safety and more efficient traffic flow, which ultimately leads to less pollution and a cleaner city.
AI represents not a replacement of human judgment but an enhancement of our capabilities — allowing us to be more proactive, more comprehensive and ultimately more effective in our safety efforts. As we continue to develop and deploy these technologies, we remain committed to creating safer environments for everyone who uses our roadways.
ABOUT VERRA MOBILITY
Verra Mobility is a leading technology company committed to improving road safety in communities across the globe. We believe transportation should be accessible, efficient and safe for all users, and we’re committed to our core mission of building safer streets and healthier communities through the use of smart transportation solutions.
Sponsor Content
