Urgent.News

What's breaking now, across thousands of outlets.

AI

17-year-old builds 170 AI glasses to help blind navigate after grandma lost sight

Seventeen-year-old Alexander Du has made waves with his creation, IRIS, an AI-powered wearable aimed at aiding the visually impaired. This device uses vibration technology to assist users in navigation and obstacle detection, providing a cost-effective alternative to pricey commercial smart glasses. His remarkable efforts were acknowledged in a ceremony at the White House, as he aspires to expand…

17-year-old builds 170 AI glasses to help blind navigate after grandma lost sight

Seventeen-year-old Alexander Du from Sewickley, Pennsylvania, has been hailed as a 2026 Presidential AI Challenge National Champion for his creation, IRIS. This AI-powered wearable device is designed to aid blind and visually impaired individuals in navigating their surroundings. The innovative glasses were inspired by Du's own grandmother, whose sight loss made navigating the world a daily struggle.

His grandmother would hold his arm tightly while crossing streets, effectively acting as her guide. This inspired Du to develop a solution that would give his grandmother—and others like her—more independence.

IRIS is a lightweight and discreet pair of glasses that cost only $170, a fraction of the price of commercial alternatives that can reach up to $6,000. The device uses AI to create a real-time 3D map of the environment and guides users around obstacles through vibration feedback. The glasses operate offline, which makes them accessible even in areas with limited connectivity. A mobile app handles navigation, while a built-in AI assistant can transcribe text and describe scenes upon voice command.

The invention faced significant challenges during its development, as Du spent late nights teaching himself AI and building IRIS from scratch. He even had to cold-email professors at the University of Pittsburgh to gain access to lab space. Despite multiple failures, Du persisted, ultimately creating a device that reduces collision accidents by 80% compared to traditional white cane usage.

Sadly, Du's grandmother passed away before he could complete the project. However, he remains committed to making IRIS accessible to millions of visually impaired individuals worldwide. He expressed that his journey taught him that when there's no path, you must create one, step by step. Du's efforts were recognized by First Lady Melania Trump at the White House during the Presidential AI Challenge awards ceremony.

Written by urgent.news from Times of India's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at timesofindia.indiatimes.com →

More in AI

I Tried to Build a Bengali Voice Dialer for Android. Here Is What Actually Happened and How I Finally got it right.

A personal project to make phone calls easier for an elderly, non English-speaking parent that works entirely offline, and the long road through Whisper models, sherpa-onnx, frame-size crashes, and…

  • Elderly, non-English-speaking mother inspired personal project
  • Turbo variant of large-v3 Whisper model most successful on Mac
  • Bengali-specific model in sherpa-onnx release assets improved accuracy

Building an evidence-first multi-agent system: 720 paired missions, rollback, and strict claim boundaries

I am an independent R&D developer building SSI V5 , an experimental multi-agent architecture focused on evidence, provenance, consolidation, and rollback. The implementation itself is proprietary.

  • SSI V5 multi-agent system focuses on evidence, provenance, consolidation, and rollback capabilities
  • Six agents (ISKRA1-ISKRA6) used in experiments with 720 paired missions
  • System passes training sequence (S1-S10) and completes drone, humanoid tasks with 100% success

Your Agent's Memory Is an Attack Surface

In Post 3, the behavioral difference tracked the memory graph, not the substrate. When I swapped one individual's learned associations into an identical brain, the subject did not just get confused.

  • Agent's memory can be compromised, turning it into an attack surface.
  • Memory graph, not substrate, shapes agent's behavior based on bias current.
  • Proposed solution: provenance of experience to differentiate earned from authored memory.

More from Sunday 20 September →