As America’s infrastructure faces increasing strain from aging roads and deferred maintenance, cities and states are strategically deploying artificial intelligence (AI) to identify critical hazards and prioritize essential repairs. This innovative approach helps to reverse alarming trends, such as the recent spike in traffic fatalities observed in Hawaii, by automating inspections and providing actionable insights into road conditions.

Hawaii officials have launched an ambitious initiative, giving away 1,000 dashboard cameras as part of their “Eyes on the Road” campaign. These dashcams, developed by Nextbase, utilize AI to automatically inspect crucial infrastructure elements such as guardrails, road signs, and pavement markings. The system can instantly differentiate between minor issues and emergencies, flagging problems that require immediate attention from maintenance crews. This real-time, continuous monitoring represents a significant leap from traditional, less frequent inspection methods. Roger Chen, a University of Hawaii associate professor of engineering involved in the program, highlighted the unique challenges faced by the state, including shipping constraints, limited space, and complex topography, which make maintaining outdated roadway infrastructure particularly difficult. The program, which was initially piloted on service vehicles in 2021 before being paused due to wildfires, also monitors street debris and faded lane lines, but particularly emphasizes the detection of damaged guardrails. Mark Pittman, CEO of Blyncsy, a company integrating dashcam feeds with mapping software, noted that the system analyzes all guardrails daily. The urgency of this focus is underscored by tragic events, such as a $3.9 million settlement with a family whose loved one died after hitting a guardrail damaged 18 months prior but never repaired, and Hawaii’s record of 106 traffic fatalities by October 2025, surpassing the total for all of 2024.

Across the continent, San Jose, California, has demonstrated early success by mounting cameras on street sweepers, which correctly identified potholes with a 97% accuracy rate. This effort is now being expanded to parking enforcement vehicles. Mayor Matt Mahan, a former tech startup founder, advocates for a shared AI database, believing that cities contributing their imagery to a common platform would significantly enhance the system's effectiveness. This allows the AI to recognize road problems it has encountered before, even if in a different location, accelerating problem identification. To facilitate this, San Jose officials helped establish the GovAI Coalition, which launched in March 2024 to enable governments to share best practices and, eventually, data. Member governments include various local entities in California, Minnesota, Oregon, Texas, and Washington, as well as the state of Colorado.

Not all AI solutions for road safety rely on cameras. Massachusetts-based Cambridge Mobile Telematics offers StreetVision, a system that analyzes cellphone data to identify risky driving behaviors. This technology works with state transportation departments to pinpoint specific road conditions contributing to these dangers. Ryan McMahon, the company’s senior vice president, recounted an instance where StreetVision detected aggressive braking on a particular road in Washington, D.C., leading to the discovery that a bush was obstructing a stop sign. The solution, in this case, was as simple as a pair of garden shears. Texas, with its extensive roadway lane miles, has integrated StreetVision alongside various other AI tools into a massive plan to improve safety. These tools have been instrumental, for example, in scanning 250,000 lane miles to efficiently identify old street signs that were long overdue for replacement, a task previously daunting with paper work orders.

Experts in AI-based road safety view current advancements as a foundational step towards a future dominated by autonomous vehicles. Mark Pittman of Blyncsy predicts that within eight years, nearly every new vehicle, regardless of whether it’s driverless, will be equipped with a camera. This future necessitates a shift in how infrastructure is conceived and managed. Departments of transportation and city agencies are now tasked with building infrastructure that caters to both human and automated drivers, bridging this evolving divide to ensure universal road safety and efficiency.