This ‘adversarial’ pattern can prevent surveillance cameras from detecting you
By Imani Sutton ·
The narrative we are currently drowning in—the one that promises a clean, smart, frictionless future—is predicated on total visibility.
The Algorithm’s Blind Spot is Our Only Leverage
The narrative we are currently drowning in—the one that promises a clean, smart, frictionless future—is predicated on total visibility. It assumes the camera always works, the algorithm never fails, and that our bodies and clothes will remain predictable data points for profit-making systems. This week’s reporting proves otherwise. Security researcher Bill Swearingen has engineered computer-generated patterns capable of hiding people, faces, and vehicles from detection by surveillance cameras. His project, noRecognition, is a blueprint for opting out of being tracked.
The concept isn't new; theguardian.com notes that Rachele Didero started designing adversarial clothing back in 2018. But the recent developments have moved this from niche counterculture to demonstrable threat mitigation. Swearingen ran 31 million tests, and his model defeated eleven open-source detection algorithms—including software powering Flock license plate readers, Axon body-worn cameras, and Clearview AI. The patterns don't block video footage; they scramble the camera’s ability to identify objects or faces, preventing the detection alerts that are supposed to be infallible.
When Fashion Becomes a Firewall Against State Power
This isn't just an academic curiosity for hackers at Black Hat. It is becoming textile engineering. mozillafoundation.org details how Italian fashion studio Cap\_able translates adversarial patterns into knit structures, creating garments that confuse AI by lowering its classification confidence or forcing it into the wrong categories. Similarly, AntiAI Fashion tests products against Google’s TensorFlow API, offering clothing designed to disrupt algorithmic tracking.
The sheer scale of this surveillance apparatus is staggering. Video surveillance enhanced by artificial intelligence is projected to grow from a $3.90bn industry in 2024 to $12.46bn by 2030. The data collection is not benign; it’s about building what the ACLU calls "a centralized dossier on everyone." We are talking about systems that collect and store massive amounts of data, allowing tracking of emotions, wardrobe, and daily habits.
The fact that these physical garments—whether they confuse facial recognition into seeing giraffes or use infrared-blocking materials to diffuse thermal signatures—are becoming available for purchase is a profound statement: the resistance has been commodified.
The Pattern of Fear Is Always Repeatable
What we are witnessing is not merely a technological arms race between pattern and pixel; it is a cyclical struggle over who gets to define "normal" behavior, and thus, who can be controlled. This dynamic echoes the mechanisms of McCarthyism. In both cases—the 1950s Red Scare and today's algorithmic surveillance boom—a powerful institutional authority attempts to define and categorize individual identity or behavior. The accusation is not based on observable facts but on the potential for deviation, leading to the suppression of dissent through a manufactured fear of detection.
The shared mechanism is clear: the promise of perfect knowledge used as a tool of exclusion. Whether it was labeling an individual's associations with "communist influence" or flagging their face in a crowd using Palantir systems, the result is the same—the curtailment of civil liberty under the guise of necessary security.
The market-driven pursuit of surveillance technology and the subsequent emergence of anti-surveillance fashion are not signs of progress; they are symptoms of an infrastructure built on systemic control. The people who design these patterns, whether in a lab or a knitwear studio, are doing more than just fooling algorithms—they are forcing us to confront the fact that our right to anonymity is not a luxury item, but a fundamental necessity for self-determination.
The moment we accept that every movement, every purchase, and every passing face must be logged, categorized, and sold as data, we have already ceded something irreversible. We cannot afford to treat privacy like an expensive niche accessory; it is the foundational infrastructure of freedom itself.