Kevin Tang, a 13-year-old innovator, has won first prize at the 2025 3M Young Scientist Challenge for developing FallGuard, an AI-powered fall detection system designed to protect elderly adults. The young scientist was awarded $25,000, which he has already partially reinvested into improving his groundbreaking project.
Tang’s motivation stemmed from personal tragedy when his grandmother fell in his kitchen and wasn’t discovered immediately, resulting in permanent brain damage. After learning that a friend’s grandparent experienced a similar incident that went unnoticed for an entire day, Tang felt compelled to create a solution for the millions of older adults who suffer from falls annually.
FallGuard uses artificial intelligence to detect falls in real-time and immediately sends alerts to family members’ phones through a dedicated mobile app. Unlike wearable devices that require charging and remembering to put on, FallGuard operates via a camera connected to a computer that can be mounted on a wall. The system can detect both sudden falls and extended periods of lying down, providing comprehensive monitoring without generating messaging fees or requiring cellular carriers.
The technology behind FallGuard leverages MediaPipe, a Google-developed AI library, which maps key points on a person’s body. Tang developed a two-stage fall detection algorithm that analyzes posture and movement over time using bounding boxes—a common computer vision tool that tracks how body proportions change from standing to lying down. If the AI detects a lay-down event, it examines the previous second to check for sudden velocity drops, distinguishing actual falls from intentional lying down.
Privacy protection is built into the system’s core design—no video is recorded or uploaded. A single FallGuard device can be linked to multiple phones, allowing several caregivers to receive simultaneous alerts. Tang is currently working to expand the system so one device can support multiple cameras throughout a home, eliminating the need for multiple computers.
Since winning the challenge, Tang has received interest from approximately 500 families. He used part of his prize money to purchase a MacBook to code the FallGuard app for computers, enabling people to convert their own devices into FallGuard systems. The project was developed with mentorship from Mark Gilbertson, a robotics and AI specialist at 3M, who praised Tang’s emotional connection to the project and his dedication to improving people’s quality of life.
Key Quotes
A few years ago, my grandma sadly fell in my kitchen, and nobody noticed immediately. It was still too late, since she was left with permanent brain damage.
Kevin Tang explained to Business Insider the personal tragedy that inspired him to create FallGuard, demonstrating how personal experience can drive meaningful technological innovation in the AI space.
You can just place [the camera] on the wall, and it works all the time. No video is recorded or uploaded, which helps protect privacy.
Tang emphasized FallGuard’s ease of use and privacy-first design, addressing two critical concerns in AI-powered home monitoring systems—convenience and data security.
This man who was trying really hard to take care of his wife, but he was deaf, so he wouldn’t hear his wife fall. This invention will just really change our lives and quality of living.
Mark Gilbertson, Tang’s mentor at 3M, shared feedback from one of the 500 families who expressed interest in FallGuard, illustrating the real-world impact of accessible AI technology for caregivers with disabilities.
I’m really proud of how much my project evolved from the very start. I just kept working until I had a final product.
When asked what he’s most proud of, Tang focused on the iterative development process rather than the accolades, demonstrating the mindset of continuous improvement that drives successful AI innovation.
Our Take
Tang’s FallGuard represents a compelling example of how AI can democratize healthcare monitoring and address critical gaps in elder care. What’s particularly noteworthy is his focus on privacy-by-design and accessibility—two areas where many commercial AI products fall short. The use of computer vision and real-time analysis without cloud storage or video recording demonstrates that effective AI solutions don’t require sacrificing privacy. His two-stage detection algorithm that distinguishes intentional lying down from falls shows sophisticated thinking about reducing false positives, a common challenge in AI monitoring systems. The fact that a 13-year-old could leverage existing AI libraries like MediaPipe to create a functional, potentially life-saving product also highlights how accessible AI development tools have become, potentially inspiring a new generation of young innovators to tackle pressing social problems with artificial intelligence.
Why This Matters
Related Stories
- Artificial Intelligence (AI) in Healthcare Market Outlook 2022 to 2028: Emerging Trends, Growth Opportunities, Revenue Analysis, Key Drivers and Restraints
- Microsoft AI CEO’s Career Advice for Young People in the AI Era
- Harvard Students Behind Facial Recognition AI Eye Construction Industry
Source: https://www.businessinsider.com/13-year-old-won-25000-for-ai-fall-detection-device-2025-12