Thirteen-year-old Kevin Tang has won $25,000 for his AI-powered fall detection device, FallGuard, which aims to help elderly individuals and their families respond quickly to accidents. The project earned him first place in the 2025 3M Young Scientist Challenge, a national competition that recognizes young innovators in science and technology. Tang’s motivation came from a personal experience. Years ago, his grandmother fell in his kitchen and wasn’t discovered for some time, resulting in permanent brain damage. Later, he learned a friend’s grandparent had also fallen, and the family only found out the next day because they lived far away. These events inspired him to create a solution that could prevent such tragedies. FallGuard uses a camera connected to a computer to monitor for falls in real time. The system leverages AI to detect when someone has fallen or has been lying down for an extended period. It then sends an instant alert to designated family members through a mobile app. Unlike wearable devices that require charging and can be forgotten, FallGuard works continuously once installed. The camera is mounted on a wall and does not record or upload video, preserving user privacy. The device uses MediaPipe, a Google-developed AI library, to map human body points on screen. Tang developed a two-stage algorithm that analyzes posture and movement over time using bounding boxes—visual markers that track changes in body proportions from standing to lying down. If a sudden drop in motion is detected, the system determines whether it’s a fall or intentional lying down, reducing false alarms. Currently, one computer can support only one camera, and the device only works within the camera’s field of view. Tang is working on expanding the system so a single device can manage multiple cameras throughout a home, eliminating the need for multiple computers. He received mentorship from Mark Gilbertson, a robotics and AI specialist at 3M, who helped with technical setup and design. Gilbertson was impressed by Tang’s emotional connection to the project and his dedication. Since winning, Tang has received interest from around 500 families. One story that stood out to him was a deaf man trying to care for his wife—FallGuard could help him know if she fell, even if he can’t hear. Tang used part of his prize money to buy a MacBook to develop the FallGuard app, which allows anyone to turn their existing computer into a monitoring system. The app is compatible with most standard computers. When asked what he’s most proud of, Tang didn’t focus on the award or recognition. Instead, he highlighted the journey—from a tripod and camera to a sleek, functional device with a downloadable app. “I just kept working until I had a final product,” he said, pointing to the device on the wall behind him. For him, the real achievement is seeing his idea evolve into something that could truly improve lives.