News

Cybersecurity Researcher Develops Car Wrap That Scrambles Flock Surveillance Cameras

August 17, 2026

A leading cybersecurity researcher has developed a vehicle wrap designed to scramble automatic license plate readers and prevent surveillance systems from recognizing a car, offering drivers a potential way to evade increasingly controversial tracking technology.

Cybersecurity researcher Bill Swearingen demonstrated the pattern at this year’s DEF CON conference in Las Vegas, where a wrapped 2009 Toyota Yaris successfully defeated surveillance detection during its first public test.

The wrap’s pattern confuses the cameras, making them unable to detect the vehicle or read its license plate.

Researcher Says Pattern Can Confuse Surveillance Systems

Swearingen created the computer-generated pattern using a reinforcement learning model trained to identify designs capable of evading computer vision systems.

Once enlarged and applied across a vehicle as a vinyl wrap, the pattern is designed to prevent surveillance cameras from recognizing the object as a car.

Despite the Toyota Yaris remaining easily identifiable as a passenger vehicle to the human eye, Swearingen said the test worked.

“We proved it was effective,” Swearingen told TechCrunch.

He said the project was intended to give people a way to resist constant surveillance.

“Privacy is a fundamental right,” Swearingen said, adding that he wanted to give Americans a way to “opt out of being tracked.”

According to TechCrunch, the pattern confused all 11 open-source detection algorithms Swearingen tested, along with Flock Safety license plate readers and Axon body-worn police cameras.

Flock Faces Growing Public Backlash

The experiment comes as Flock Safety faces mounting opposition to its expanding network of automatic license plate readers across the United States.

The company has reportedly lost dozens of contracts with cities as public support for widespread surveillance declines.

Flock also developed plans to use dashcams installed in Uber and Lyft vehicles to monitor other cars, according to 404 Media.

Citizens opposed to the cameras have increasingly resorted to disabling or damaging them, including cutting them down, blocking their view, and using other forms of sabotage.

Swearingen’s project offers a different approach by attacking the surveillance technology at the software level rather than physically interfering with the cameras.

Clothing Could Also Defeat Computer Vision

Swearingen has expanded the concept through his noRecognition crowdsourcing campaign.

He has developed T-shirts and hoodies featuring algorithmically generated patterns intended to confuse computer vision systems.

Eventually, he hopes to sell full-vehicle skins using the same technology.

The project is part of a wider movement involving companies and researchers developing so-called adversarial patterns that exploit weaknesses in artificial intelligence-powered image recognition.

As license plate readers and other automated surveillance systems spread across American cities, Swearingen’s experiment demonstrates that the same artificial intelligence used to track citizens can also be manipulated to defeat the technology.