In just 20 minutes this morning, an automated license plate recognition (ALPR) system in Nashville, Tennessee, captured photos and detailed information from nearly 1,000 vehicles as they drove by. Among them: eight black Jeep Wranglers, six Honda Accords, an ambulance, and a yellow Ford Fiesta with a vanity plate. This trove of real-time vehicle data, collected by one of Motorola’s ALPR systems, is obviously accessible to law enforcement. . However, flaws discovered by security researchers have exposed live video feeds and detailed records of passing vehicles, revealing the staggering scale of surveillance enabled by this pervasive technology. More than 150 Motorola ALPR cameras have had their video feeds and data leaked in recent months. , according to security researcher Matt Brown, who first disclosed the problem in a series of YouTube videos after buying an ALPR camera on eBay and reverse engineering it. broadcasting live footage accessible to anyone on the internet, misconfigured cameras also reveal the data they collect, including car photos and license plate logs. Real-time video and data feeds do not require a username or password to access. Along with other technologies, WIRED has reviewed the video feeds from several cameras, confirming vehicle data—including the car’s make, model, and color—has been intentionally exposed. Motorola confirmed the exposure, telling WIRED that it is working with its customers to close the access. In the last decade, thousands of ALPR cameras have appeared in US cities and towns. The cameras, manufactured by companies such as Motorola and Flock Safety, automatically take pictures when they detect a passing car. The cameras and the database of data they collect are often used by the police to search for suspects. ALPR cameras can be placed on the road, on the dashboard of police cars, and even on trucks. This camera takes billions of photos of cars—including the occasional bumper sticker, yard sign, and T-shirt. , tells WIRED. The video feeds each cover one lane of traffic, with cars driving through the camera’s view. In some places, snow falls. Brown found two streams for each exposed camera system, one in color and the other in infrared. Generally, when a car passes an ALPR camera, a photo of the vehicle is taken, and the system uses machine learning to extract the text from the license plate. This is stored along with details such as where the photo was taken, the time, as well as metadata such as the make and model of the vehicle. networks, possibly by law enforcement agencies that distribute them, and even appear on the internet without authentication. “It has been misconfigured. It should not open on the public internet,” he said. to New York City. In just 20 minutes, the camera recorded the make, model, color, and license plate number of nearly 4,000 vehicles. Some cars are even captured multiple times—up to three times in some cases—as they pass through different cameras.