An example of data collected by a street activity sensor, with mapping of movement by pedestrians, cyclists, and cars at an intersection. Credit: NYC DOT

Across New York City, signposts are being fitted with a new generation of traffic sensors. These sensors count pedestrians, cyclists, buses, trucks, and cars in real time. The devices use machine vision to track how people move through streets — including speeds, turns, mid-block crossings, and potential conflict points. NYC officials say the video is immediately discarded and only anonymized movement data is kept.

The goal is to give planners a clearer picture of where street design is failing, so they can add safer crossings, better bike routes, bus improvements, and other fixes where they’re needed most.

But will it actually work?

What’s Actually Happening on the Street

This is basically a larger roll-out of a previous program by the Department of Transportation (NYC DOT). After a 2023 pilot at 20 locations, the agency plans to add sensors at roughly 80 more sites, bringing the network to about 100 locations. The promise is a richer, round-the-clock picture of how New Yorkers actually move through streets—and where the street design fails them.

NYC DOT Traffic Operations worker installing a new activity sensor in Greenpoint, Brooklyn. Credit: NYC DOT

The sensors, provided by Viva and mounted on existing street infrastructure, act like traffic counters. They can record how many people walk, bike, or drive through an area, how fast vehicles move, where people turn, and where pedestrians cross outside marked crosswalks.

That kind of information matters a lot.

A steady stream of people crossing mid-block could suggest that a new crosswalk is needed. Repeated close calls between cyclists and opening car doors could flag a corridor for a protected bike lane or parking changes, even when there are no reported accidents. A bus route that keeps getting trapped behind turning vehicles could point planners toward a dedicated lane or signal tweak.

“Safer street design starts with understanding what is actually happening on the street,” NYC DOT Commissioner Mike Flynn said in the agency’s announcement. “These high-tech sensors will help us evaluate how people are walking, biking, and driving so we can design safer streets and encourage safer behavior.”

×

Thank you! One more thing…

Please check your inbox and confirm your subscription.

Until now, much of this work has relied on people with clipboards watching intersections for short stretches of time. That method can be useful, but it misses a lot: late-night speeding, weekend patterns, seasonal changes, and near-misses that never show up in crash data. Plus, the task itsel is very challenging for humans.

“It’s not easy to be able to write down vehicle types at the same time as the number of volumes, for example,” Eric Beaton, the DOT’s deputy commissioner for transportation planning and management, told Gothamist. “These sensors provide a much richer set of data for us to work with.”

The Transparency Test

NY Traffic. Credit: Pexels

Still, the program comes with an obvious tension. Everyone wants better street-safety, but efw people want another layer of surveillance.

“There’s nothing that we ever touch or that anyone could ever touch that has anything identifying to any person or any vehicle,” Beaton added.

The agency says the technology processes video in real time, discards footage immediately, and keeps only anonymous data. The system obscures faces and license plates before any human sees the results.

Still, advocates want transparency. Jon Orcutt, a transit advocate and former DOT policy director, told Gothamist that New York has too little information on walking and cycling. “There are 20 or 25 sites where bikes are counted,” he said. “You get a very small data set from that. You know we have 6,000 miles of streets.”

“If they’re collecting this data on behalf of the public, as a taxpayer-funded agency, we deserve to know what it says and so there should be regular reporting,” Orcutt argued.

The expansion will cost about $200,000 in city funds, on top of $100,000 for the pilot, with grants covering the rest. During the pilot, human inspectors ran parallel counts, and officials said the sensor results checked out.

The next test will be whether the city turns the data into visible street changes, and whether it shares enough of that data for New Yorkers to judge the program themselves. A sensor can count a near-miss, but the public will measure success in safer crossings, better bike routes, and fewer crashes.