Science

AI Breakthrough Could Replace Rare Earth Magnets in Electric Vehicles — 67,000 New Materials Discovered

Scientists at the University of New Hampshire have used artificial intelligence to build the world's largest searchable database of magnetic materials — and in the process, discovered 25 entirely new magnets that could help end our dependence on costly, scarce rare earth elements.

The Rare Earth Problem

Magnets are everywhere in modern life — in your smartphone, your car, medical devices, wind turbines, and power generators. But today's most powerful permanent magnets rely on rare earth elements that are expensive, difficult to mine sustainably, and concentrated in a few countries (China controls roughly 60% of global production).

Despite thousands of known magnetic compounds, no entirely new permanent magnet had been identified from this pool — until AI entered the picture.

🧲 By the Numbers

  • 67,573 magnetic compounds catalogued in the new database
  • 25 new high-temperature magnets discovered
  • Published in: Nature Communications (February 2026)
  • Database: Northeast Materials Database — free and searchable

How AI Cracked the Problem

The team developed an AI system capable of reading thousands of scientific papers and automatically extracting experimental data about magnetic properties. That information was used to train machine learning models that can predict whether a material is magnetic and calculate the temperature at which it loses its magnetism (the Curie temperature).

"By accelerating the discovery of sustainable magnetic materials, we can reduce dependence on rare earth elements, lower the cost of electric vehicles and renewable-energy systems, and strengthen the U.S. manufacturing base," said Suman Itani, lead author and doctoral student in physics at UNH.

Why This Changes Everything

The implications are enormous. If even a few of these 25 newly discovered magnets prove commercially viable, they could:

  • Make electric vehicles cheaper — rare earth magnets are one of the most expensive components in EV motors
  • Secure clean energy supply chains — wind turbines could be built without geopolitically sensitive materials
  • Reduce environmental damage — rare earth mining is notoriously polluting
  • Democratize manufacturing — countries without rare earth deposits could build their own EV and renewable energy industries

💡 The Bigger Picture

This is a perfect example of AI being used for genuine, tangible good — not replacing human jobs, but doing something humans simply couldn't do at this scale: reading thousands of papers, connecting the dots, and finding materials that had been hiding in plain sight for decades.

The Northeast Materials Database is now freely available for researchers worldwide, meaning this is just the beginning. The age of sustainable magnets may be closer than anyone thought.

☕ Enjoy our positive journalism? Support Good News 24

Buy Us a Coffee

📚 Recommended Reading

Dive deeper into AI and science breakthroughs:

Affiliate links — purchases support Good News 24 at no extra cost to you.