The MTA has partnered with Google to use AI-powered smartphones to detect track defects in New York City's subway system, aiming to reduce delays and improve efficiency.
 |
| The MTA collaborates with Google on TrackInspect, using AI-driven smartphones to enhance subway track inspections and reduce disruptions in New York City's transit system. Image Courtesy: MTA |
New York City, USA — March 11, 2025:
In an innovative move to reduce delays and improve efficiency in New York City’s subway system, the Metropolitan Transit Authority (MTA) has partnered with Google on a groundbreaking pilot program. The initiative, known as TrackInspect, retrofits subway cars with Google Pixel smartphones to monitor track conditions using artificial intelligence, according to MTA, CNN.
The project, which began in September 2024, involved installing smartphones equipped with sensors and microphones on select subway cars. These devices collected over 1,200 hours of audio, 335 million sensor readings, and 1 million GPS locations, sending the data to Google’s Cloud for analysis. The AI system then analyzed these inputs for signs of potential track defects, helping to identify issues before they could disrupt service.
Rob Sarno, assistant chief track officer at the MTA, explained how the system used the collected data to detect subtle vibrations and sounds, allowing the AI to predict track problems like loose joints or damaged rails. According to Sarno, the system was able to correctly identify 92% of the defects found by human inspectors, marking the pilot as a success.
“By detecting these issues early, we can save both time and money while minimizing disruptions for commuters,” said Demetrius Crichlow, president of New York City Transit.
"How it works is the prototype sends a soundbite or noise clip showing heavy vibration or noise, and then our inspectors follow up by walking the track and verifying any issue found,” said New York City Transit Department of Subways Assistant Chief Track Officer Robert Sarno. “We then compare that with whatever we find to teach the device noise and decibel levels and then work from there. That's how we are able to instruct the prototype on what are normal sounds and vibrations, and what are not, and move along through the process.”
TrackInspect is part of a broader trend in major cities worldwide to incorporate AI into public transit systems. Cities like Chicago and New Jersey have tested similar technologies, while Beijing has introduced facial recognition for ticketing. However, the MTA’s pilot program is notable for its focus on using AI to improve track maintenance and reduce delays in the country’s largest transit system, which serves over 1 billion riders annually.
Despite the promising results, the MTA faces significant budgetary challenges. The program, which was developed at no cost to the MTA by Google Public Sector, remains in the pilot phase, and its future expansion depends on how the technology fits into the MTA’s budget and long-term plans. The MTA is continuing to explore other partnerships to improve track inspection and overall service efficiency.
The MTA’s commitment to using advanced technology for transit improvements highlights the growing role of AI in public infrastructure. While TrackInspect has shown potential, the broader adoption of such technologies across the system remains uncertain, as the MTA weighs costs against benefits.