How a 17-year-old is using OpenAI tools to make animal healthcare smarter

For most teenagers, technology is about convenience. For 17-year-old Anshul Bhatt, it became a way to solve a problem that hit close to home.

When his family dog, Max, was diagnosed with advanced arthritis, Anshul discovered how limited and expensive diagnostic tools for animals really were. By the time vets could identify Max’s condition, it was too late to reverse the damage.

“That experience made me realize how invisible early-stage diseases can be in animals,” Anshul says. “If we could spot them earlier, the outcomes would be completely different.”

With curiosity as his guide, and OpenAI’s tools as his learning partner, Anshul began exploring how artificial intelligence could help detect mobility issues in animals through movement patterns. What started as a personal project soon evolved into PawPath, an AI-powered system that uses motion sensors and machine learning to identify early signs of orthopedic and neurological disorders in dogs.

A student project with real-world purpose

PawPath uses four lightweight sensors attached to a dog’s legs, recording their motion hundreds of times per second. The data is analyzed by a machine learning model that identifies patterns linked to orthopedic and neurological conditions such as arthritis, ligament injuries, or ataxia.

Unlike high-end lab equipment, PawPath is portable and low-cost, making it practical for shelters and small veterinary clinics. Anshul is currently testing the device with large animal welfare centres, where it has already helped identify mobility issues in rescued dogs.

“This isn’t about replacing vets,” Anshul clarifies. “It’s about giving them a faster, data-backed way to decide which animals need help first.”

Learning from AI, not just building with it

Working beyond traditional resources, Anshul turned to OpenAI’s tools as his main research support. He used ChatGPT to understand complex topics like sensor fusion, motion tracking, and Kalman filters — concepts often reserved for graduate-level studies.

“I used OpenAI tools to learn and debug,” he says. “It helped me grasp difficult ideas faster, especially when I didn’t have someone to ask.”

Instead of replacing the hard work of research, OpenAI accelerated it. It became the tutor, collaborator, and coding assistant that allowed him to move from reading to building, he shares.

Expanding horizons: AI in medical imaging

PawPath wasn’t the end of Anshul’s AI journey. As part of a summer research program at MIT, he collaborated with a team at Harvard Medical School to explore how AI could reduce repetitive work in radiology.

There, he built voice-based AI agents powered by OpenAI models that could navigate radiology interfaces, automate basic image review tasks, and help reduce fatigue among clinicians. “The idea was to make AI useful for the people doing the work, not to replace them,” he explains.

Both projects, though vastly different, are rooted in the same belief that AI can extend human capability in meaningful ways, whether for doctors or for animals.

Building for impact, not recognition

For Anshul, the goal has never been awards or accolades, though they’ve come along the way – from winning the Grand Award at the Regeneron International Science and Engineering Fair to being recognized among the top 100 young researchers worldwide.

The real reward lies in using AI to make a tangible difference. “I’ve realized you don’t need a lab or a team of experts to start,” he says. “You just need curiosity, and a willingness to build.”

From diagnosing mobility disorders in dogs to automating radiology workflows, Anshul’s work shows how OpenAI is helping a new generation of young researchers learn independently, explore deeply, and build for impact.