Skip to main content

AI Agent 'Elements Claw' Discovers New Superconductors, Ushering in a New Era of Scientific Discovery

AI Takes the Lead in Superconductor Discovery

For decades, finding new superconducting materials has been a slow, painstaking process of trial and error. But that's starting to change. On July 3, a team from Alibaba's DAMO Academy, Renmin University, and the University of Chinese Academy of Sciences unveiled Elements Claw — the world's first AI agent built specifically to hunt for superconductors.

Think of it as a tireless, super-smart research assistant that never sleeps. Instead of relying on human intuition alone, Elements Claw taps into a massive database of 125 million molecules and crystal structures. It's powered by a 10-billion-parameter atomic foundation model called Elements, which can predict a material's superconducting potential with astonishing accuracy — an AUC score of 0.996, and critical temperature errors within just 1 Kelvin.

Image

From Data to Discovery in Hours

What makes Elements Claw truly revolutionary is its ability to work like a human scientist — but faster. It can read scientific literature, assess whether a material can be synthesized, design experiments, and even learn from its own findings. In a recent test, the AI sifted through 2.4 million crystal structures and identified 68,000 promising candidates — all in just 28 GPU hours. That's a job that would take human researchers years.

The team has already synthesized and verified four new superconductors, including HfZrRe4, which was designed from scratch by the AI, and three others discovered by correcting and reanalyzing existing databases: Hf21Re25, Zr4VRe7, and Zr3ScRe8. Their critical temperatures reach up to 6.5K.

A New Blueprint for Materials Science

"These results show the huge potential of AI agents in material discovery," said Rong Yu, head of Science Intelligence at DAMO Academy. To help accelerate the field, the team has released the full dataset of 2.4 million stable crystals.

Professor Huang Wenbing from Renmin University's Gaoqiang Institute of AI added that this framework could be reused for other critical materials, such as solid-state battery electrolytes, multiphase catalysts, and thermoelectric materials. "It opens a new chapter in scientific discovery," he said.

Key Points

  • Elements Claw is the first AI agent designed for superconducting material discovery.
  • It screened 68,000 candidates from 2.4 million structures in 28 GPU hours.
  • Four new superconductors have been synthesized, with critical temperatures up to 6.5K.
  • The AI uses a 10-billion-parameter atomic model with near-perfect prediction accuracy.
  • The team has released the full dataset to promote further research.