Bill Swearingen, a cybersecurity researcher, has created computer‑generated patterns that can prevent many surveillance cameras and license‑plate readers from detecting people or vehicles. Surveillance cameras increasingly use automated detection to track license plates and identify faces, raising privacy concerns. Swearingen’s project, called noRecognition, produces patterns that scramble visual data so cameras cannot trigger detection alerts. The patterns do not stop recording; they make the covered area appear invisible to detection algorithms. After more than 31 million tests, his reinforcement‑learning model can generate patterns that defeat 11 open‑source detection algorithms, including those used by Flock license‑plate readers, Axon body cameras and Clearview AI. At the Def Con conference in Las Vegas, he covered a 2009 Toyota Yaris with a new pattern and showed that a Flock camera failed to detect the vehicle, noting that the wheels remained a challenge and that video of the demo will be released later. The project is seeking crowd‑sourced funding for merchandise that incorporates the patterns, such as T‑shirts and hoodies, and aims to make the patterns high‑resolution and aesthetically appealing. Swearingen is keeping his strongest patterns private to prevent camera manufacturers from countering them, but says the model continues to improve. "Privacy is a fundamental right," Swearingen said. "The patterns let people opt out of being tracked."