Researcher Unveils Pattern That Evades Surveillance Cameras
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- Bill Swearingen developed a computer-generated pattern through 31 million tests that prevents common surveillance systems from detecting people or vehicles covered by the design.
- Swearingen demonstrated the pattern at Def Con by successfully hiding a Toyota Yaris from a Flock license plate reader, proving real-world effectiveness despite challenges with wheel visibility.
- noRecognition, Swearingen’s project, uses reinforcement learning to evolve patterns that defeat detection algorithms, including those used by Flock, Axon, and Clearview AI, across 11 open-source models.
- Swearingen described his model as 'teaching itself how to paint,' refining patterns after each failure until they consistently avoid detection, with new designs generated every minute.
- The noRecognition project launched a crowdfunding campaign to distribute patterned merchandise like T-shirts and hoodies, aiming for both high-resolution functionality and aesthetic appeal.
- Swearingen is withholding his strongest patterns from public release to prevent camera manufacturers from adapting to and neutralizing them, while continuing to improve the system.
Why it matters: Surveillance-dependent law enforcement and private monitoring systems could lose reliability if evasion tools become widespread, undermining automated detection investments. With 31 million iterations already completed and patterns evolving autonomously, the technology presents a scalable challenge to current public monitoring infrastructure without requiring illegal access or hardware modification.
Ask SkimNews
