VDOT Bets on AI to Cut Costs and Keep Virginia Roads Smooth

With highway construction costs surging 68% nationwide since 2020, Virginia’s transportation officials are turning to artificial intelligence to improve how the state estimates costs and manages pavement upkeep.

The move comes as the Virginia Department of Transportation (VDOT) faces increasing financial strain from rising material and labor prices, supply chain disruptions, fuel costs, and unpredictable weather events — all of which have made forecasting and budgeting more difficult.

“It is a great time for VDOT to investigate the potential for Artificial Intelligence to enhance the way we make decisions given our ongoing commitment to improving our data collection and management,” said Cathy McGhee, chief deputy commissioner for VDOT, in a statement to the Mercury.

McGhee emphasized that with the cost of construction climbing, AI-driven insights could help maximize VDOT’s investments in maintenance and repairs.

“Increasing construction costs make it even more important than ever to have data-driven decisions with respect to our investments,” she said. “Although we believe that there are many areas where AI could be beneficial, these initial pilots will focus on cost estimation and pavement management, two areas where cost impacts could be significant.”

Virginia maintains one of the largest state-controlled roadway networks in the country, covering nearly 60,000 miles of roads, in addition to at least 10,000 miles in cities and towns. Over the past decade, 3,200 additional lane-miles have been added to VDOT’s maintenance list.

Pavement resurfacing alone has risen as much as 45% since 2019, and VDOT is responsible for a growing list of maintenance projects, including patching potholes, resurfacing roads, and mowing roadside vegetation.

During a recent presentation to the Commonwealth Transportation Board’s Subcommittee on Innovation, McGhee and VDOT staff outlined their goals for the AI pilots, emphasizing that the technology  could provide deeper insights into the lifecycle of pavement beyond existing agency data.

Gov. Glenn Youngkin’s administration and state lawmakers have been actively preparing Virginia for AI integration across multiple sectors, including communications, health care, and security. However, AI’s potential for misuse in politics has also led to increased scrutiny and regulatory efforts.

Last year, Youngkin signed an executive order establishing standards and guidelines for AI use in government, forming  a task force to help lawmakers create guardrails for responsible AI implementation.

The governor is now considering  some AI-related legislation passed by the General Assembly, including House Bill 2094, introduced by Del. Michelle Maldonado, D-Manassas, which would create requirements for developing, deploying and using high-risk AI systems, along with civil penalties for noncompliance.

Other proposed measures would limit social media use for minors under 16, create enhanced penalties for AI-generated deepfakes used in fraud, slander, libel or defamation, and require political ads to disclose when artificial intelligence has been used.

As Virginia awaits Youngkin’s final decisions on AI legislation, Maldonado said she is particularly interested in the results from VDOT’s AI-driven pilots.

“I think it’s a very good start,” Maldonado said. “I think it’s a smart way to use AI, which helps to augment versus replace the human element, and provides better data analysis and hopefully traffic flow.”

by Nathaniel Cline, Virginia Mercury

Virginia Mercury is part of States Newsroom, a nonprofit news network supported by grants and a coalition of donors as a 501c(3) public charity. Virginia Mercury maintains editorial independence. Contact Editor Samantha Willis for questions: info@virginiamercury.com.