Shirt Aims to Foil Recognition Systems

American researcher Bill Sueringen proposed a project called "noRecognition" aimed at reducing the ability of AI-powered surveillance systems to track people and vehicles by confusing facial recognition algorithms. The project involved designing printed patterns that do not prevent cameras from capturing images but cause disruptions in image analysis, leading recognition systems to fail in accurately classifying targets. Over the course of a year, approximately 31 million tests were conducted to develop effective patterns, which were tested against systems such as vehicle license plate readers and facial recognition cameras, successfully disabling them. The technique was showcased at the "Def Con" conference and successfully fooled a surveillance system during a test drive of a covered car in front of AI-based cameras, despite some technical challenges, like not fully obscuring the wheels. Notably, Sueringen has begun selling his designs through products like T-shirts and apparel, with plans to develop custom covers for vehicles. However, the effectiveness of the technique may diminish as AI algorithms are updated and new systems emerge. The project highlights an effort to balance technological advancement with individual privacy rights.

articleDetails.fontSize.sizeحجم النص

تنويه: هذا ملخص تم إنشاؤه بواسطة الذكاء الاصطناعي

عرض المقال بالكامل