New York transit expands AI-powered technology trials

New York transit agencies are expanding their use of AI and data-driven technologies as they seek to modernise ageing infrastructure, streamline workflows and improve operational efficiency, with 18 companies selected for the latest challenge cycle of the Transit Tech Lab.

The latest cohort reflects a growing shift toward AI-enabled tools, with more than half of participating companies deploying AI or autonomous technologies across areas including predictive maintenance, transit planning, digital twins and infrastructure inspection.

“There are two major themes: extending the reach of inspection and condition assessment work, and reducing administrative burden on back-office workflows so staff can focus on higher-value work,” Stacey Matlen, Senior Vice President of the Partnership Fund for New York City, which operates the Transit Tech Lab, told Cities Today.

The comments point to a growing focus on AI applications that help agencies expand inspection capacity and improve operational decision-making without removing human oversight.

“On infrastructure, the appetite is for tools that give existing inspectors and engineers more coverage and better data to work with,” she said. “Continuous camera and sensor data flagged by AI means an inspector can prioritise where to spend their time and catch issues between scheduled inspections.”

The second focus area centres on modernising workflows and making better use of operational information held across disconnected systems.

“On workflows, the appetite is for AI that takes the most repetitive parts of administrative work, like drafting boilerplate or surfacing context from scattered systems, and lets staff spend their time on the parts of the job that require expertise and judgment.”

Among the selected companies are AI procurement platform Hazel, transit planning company Ontra Mobility and contextere, which connects data across existing systems to provide operational recommendations to frontline workers.

Despite increasing interest in AI, Matlen said agencies remain cautious around oversight and security concerns.

“Hesitation tends to cluster around AI making consequential decisions without enough human oversight, including the risk of models being compromised by bad actors, and around protecting sensitive operational and rider data,” she said.

The Transit Tech Lab uses controlled proof-of-concept environments and phased testing processes designed to demonstrate value while limiting risk. Matlen said successful scaling depends on technologies addressing existing operational needs rather than introducing standalone capabilities.

“The technology solves a problem the agency was already trying to solve,” she said. “Solutions scale when they map to a known operational pain point with a clear plan for how to address it.”

The eighth annual programme, run by the Transit Tech Lab alongside agencies including the Metropolitan Transportation Authority (MTA), Port Authority of New York and New Jersey, NYC Department of Transportation, NJ TRANSIT and NYC Department of Design and Construction, will see companies test technologies through eight-week proof-of-concept projects. Since launching in 2018, more than 1,000 companies have applied to participate, with 81 technologies tested and 22 solutions commercially scaled.

Main image: William87 | Dreamstime.com