AI Ends Century-Long Superconductivity Race, Identifies Four New Materials in 28 Hours
In physics, the search for superconductors has long been seen as a century-long pursuit of the 'Holy Grail.' Recently, this traditional field gained a powerful AI collaborator — an agent named 'ElementsClaw' (Elemental Crab), built specifically to discover superconducting materials. Within just 28 GPU hours, the system scanned 2.4 million stable crystals and identified 68,000 potential superconductors, achieving research efficiency far beyond centuries of human exploration.
For years, the search for superconductors relied on a 'cookbook' approach — testing different element combinations through trial and error. This method had a very low success rate and often depended on accidental discoveries. To change that, the research team developed a 'convergent expertise' architecture, equipping ElementsClaw with unique research capabilities: a built-in 1-billion-parameter geometric deep graph neural network that accurately interprets three-dimensional crystal structures, combined with a large language model that allows the AI to autonomously read literature, access data, and assist in decision-making.

This AI system is not just a powerful tool for 'finding a needle in a haystack' — it also has self-evolution capabilities. During experimental validation, researchers successfully synthesized four entirely new superconductors previously unknown to humans. Each discovery showcased different logical paths taken by the AI, from re-evaluating forgotten database structures, to correcting past computational errors, to generalizing from structural motifs. ElementsClaw demonstrated a leap from 'assisted judgment' to 'active design' in materials science.
Although the critical temperatures of these newly discovered materials have not yet reached room temperature, their core value lies in validating the application of AI agents in this field. Compared to the natural hit rate of about 3% for superconducting materials, ElementsClaw's recommendation accuracy improved by an order of magnitude.
Related article
Nvidia H200 China deal survives Trump-Xi summit, but not as expected
President Trump traveled to Beijing with Jensen Huang joining the delegation at the eleventh hour, departing two days later with vague remarks about potential chip export developments. No deal materialized. Since Trump first authorized sales in Decem
How to fix Core Web Vitals for SEO ranking?
Creating visually appealing assets is crucial in game development. Pixel art, with its charming retro aesthetic, remains a popular choice. With the advent of AI tools, generating pixel art has become more accessible than ever. This guide explores how
TechCrunch Disrupt 2026 breakout session agenda unveiled
TechCrunch Disrupt 2026 arrives at Moscone West in San Francisco, October 13–15, uniting 10,000+ founders, investors, operators, and tech leaders for three days of insight, networking, and forward-looking dialogue. Register now to save up to $100 on
Related Special Topic Recommendations
Comments (0)
0/500
In physics, the search for superconductors has long been seen as a century-long pursuit of the 'Holy Grail.' Recently, this traditional field gained a powerful AI collaborator — an agent named 'ElementsClaw' (Elemental Crab), built specifically to discover superconducting materials. Within just 28 GPU hours, the system scanned 2.4 million stable crystals and identified 68,000 potential superconductors, achieving research efficiency far beyond centuries of human exploration.
For years, the search for superconductors relied on a 'cookbook' approach — testing different element combinations through trial and error. This method had a very low success rate and often depended on accidental discoveries. To change that, the research team developed a 'convergent expertise' architecture, equipping ElementsClaw with unique research capabilities: a built-in 1-billion-parameter geometric deep graph neural network that accurately interprets three-dimensional crystal structures, combined with a large language model that allows the AI to autonomously read literature, access data, and assist in decision-making.

This AI system is not just a powerful tool for 'finding a needle in a haystack' — it also has self-evolution capabilities. During experimental validation, researchers successfully synthesized four entirely new superconductors previously unknown to humans. Each discovery showcased different logical paths taken by the AI, from re-evaluating forgotten database structures, to correcting past computational errors, to generalizing from structural motifs. ElementsClaw demonstrated a leap from 'assisted judgment' to 'active design' in materials science.
Although the critical temperatures of these newly discovered materials have not yet reached room temperature, their core value lies in validating the application of AI agents in this field. Compared to the natural hit rate of about 3% for superconducting materials, ElementsClaw's recommendation accuracy improved by an order of magnitude.
Nvidia H200 China deal survives Trump-Xi summit, but not as expected
President Trump traveled to Beijing with Jensen Huang joining the delegation at the eleventh hour, departing two days later with vague remarks about potential chip export developments. No deal materialized. Since Trump first authorized sales in Decem
How to fix Core Web Vitals for SEO ranking?
Creating visually appealing assets is crucial in game development. Pixel art, with its charming retro aesthetic, remains a popular choice. With the advent of AI tools, generating pixel art has become more accessible than ever. This guide explores how
TechCrunch Disrupt 2026 breakout session agenda unveiled
TechCrunch Disrupt 2026 arrives at Moscone West in San Francisco, October 13–15, uniting 10,000+ founders, investors, operators, and tech leaders for three days of insight, networking, and forward-looking dialogue. Register now to save up to $100 on





Home






