Google DeepMind AI predicts about 48,000 new materials
Make us a preferred source
Make us a preferred source on Google
The news: Google's artificial intelligence research laboratory DeepMind claims it has predicted the structure of about 48,000 new materials.
Robots are being used to test the integrity of these materials in a US university laboratory.
The context: The claims were made in research published in the science journal Nature. The materials could be used to make better batteries, solar panels and computer chips.
DeepMind's artificial intelligence was trained on data from the Materials Project. The international research group uses findings based on about 500,000 materials.
The A-Lab at the University of California is using artificial intelligence robots to analyse and create new materials.
What they said: "This is like ChatGPT for materials discovery," Cornell University AI for Science Institute co-director Carla Gomes said.
"A lot of the technologies around us, including batteries and solar cells, could really improve with better materials," Google DeepMind's Ekin Dogus Cubuk said.
The source: Nature
