Alexander Du, a 17-year-old student from Sewickley, Pennsylvania, has been named a 2026 Presidential AI Challenge National Champion for inventing IRIS, an AI-powered wearable designed to help blind and visually impaired people navigate their surroundings. The teen built the $170 system after watching his grandmother struggle to navigate the world differently after losing her sight. The lightweight, discreet glasses he developed act as a virtual guide dog, helping users detect obstacles and navigate around them using physical vibrations. An invention born out of pure necessity Alexander Du never planned to become an inventor. But then he saw his grandmother struggle after losing her sight. He saw the toll of blindness firsthand. She held his arm tightly as they crossed streets. He became her guide, narrating curbs, sidewalk cracks, everything she could no longer see. “It can be detrimental to people’s lives,” Du told Trib Live. “That inspired me. At home, I got more into it. I thought, ‘What if there’s a way to give my grandmother her independence back?’”Du realised that all the help his grandmother got from the family was still inefficient. He wanted her to have her independence. A traditional walking stick was still insufficient. There were other technologies, such as smart glasses, but they were still expensive. So he decided to build one called IRIS (Intelligent Responsive Imaging System), a pair of lightweight glasses that cost $170 and work like a virtual guide dog. The device maps its surroundings in real-time 3D, guides users around obstacles through vibration feedback, and operates completely offline. How IRIS sees what others can’t The glasses Du built use AI to create a three-dimensional map of the environment. They are constantly processing what is ahead and around the wearer. When an obstacle appears, the glasses don’t just alert the user; they steer them through vibrations. Think of it like a physical nudge in the right direction. The results speak for themselves. The system reduces collision accidents by 80% compared with a traditional white cane.This system comes paired with a mobile app that handles the navigation. GPS guides users between destinations. There is a built-in AI assistant that transcribes text and describes scenes on voice command. Everything runs on the device; there is no need for it to be connected to the cloud or internet. And for someone navigating the world independently, these simple things really matter. The price point of the glasses also matters. Commercial alternatives cost upwards of $6,000 and are out of reach for many visually impaired people worldwide. Du has built something powerful for a fraction of that cost.But none of this was easy for the 17-year-old. He spent late nights teaching himself AI and building IRIS from scratch. Over those months, he cold-emailed professors at the University of Pittsburgh. That was when they granted him access to loaner lab space. He failed multiple times before finding success.Unfortunately, Du’s grandmother passed away before he completed his work on IRIS. But he is committed to making the system accessible to millions of others. “In developing IRIS, I learned that when the path doesn’t exist you make it yourself — one paper, one cold email, one failed solder joint at a time.”Meanwhile, Du’s efforts also grabbed the attention of First Lady Melania Trump. The teen was recognised at the White House during the Presidential AI Challenge awards ceremony, an event that honoured young Americans using AI to develop technology.“My experience at the White House was truly amazing. I’m still trying to process it. Getting to shake the first lady’s hand, and getting the recognition for the work I’ve spent the past months on has been really rewarding,” Du told the outlet.Du has also won the Barron Prize 2026. Source link Post Views: 7 Post navigation ‘When do I blow entire nation up?’ Trump’s ultimatum after Iran warns of ‘decisive war’ Meet Ameya Kharade, the 17-year-old Nashua student whose $1,800 brain-reading system aims to turn brain activity into speech at implant-like speeds without surgery