Microsoft research has launched a new powerful AI system today that creates new devices with some of the desired properties, which can accelerate the development of better batteries, energy efficient solar cells and other complex technologies.

The system, called MatterGenit represents a major change in the way scientists discover new things. Instead of sifting through millions of existing products – a traditional process that can take years – MatterGen directly creates new materials based on your needs, similar to how AI image creators create images from verbal descriptions.

“Creative products provide new ideas for tool design by directly creating new products that address the challenges you need,” said Tian Xie, senior research manager at Microsoft Research and executive editor. research published today in Nature. “This represents a major step forward in the development of a world-class equipment manufacturer.”

How Microsoft’s AI engine works is different from traditional methods

MatterGen uses a special type of AI called a spread the pattern – similar to those behind image generators such as DALL-E – but modified to work with three-dimensional lenses. It randomly refines atoms into stable, useful substances that meet the requirements.

The results surpass previous methods. According to the research paper, the devices created by MatterGen are “twice as fast to be innovative and stable, and more than 15 times closer to the minimum energy” compared to previous AI methods. This means that the products produced can be useful and can be manufactured.

In one interesting demonstration, the group collaborated with scientists in China Shenzhen Institutes of Advanced Technology to create new things, Image of TaCr2O6which MatterGen created. Real-time data is closely related to AI predictions, validating the application of the system.

Real-world applications could revolutionize energy storage and computing

This system is very popular because of its flexibility. It can be “fine-tuned” to create devices with other properties – from crystal materials to desired electrical or magnetic properties. This can be very important in the production of special industrial equipment.

The consequences can be huge. New materials are essential for the advancement of energy storage technologies, semiconductor design and carbon capture. For example, better battery systems could accelerate the transition to electric vehicles, while solar cell systems could make renewable energy more affordable.

“In terms of industry, the potential here is huge,” Xie explained. “Human development always depends on new things. If we can use artificial intelligence to create more efficient tools, it can improve the performance of industries such as energy, healthcare and others. “

Microsoft’s open source strategy aims to accelerate scientific discoveries

Microsoft he has released MatterGen’s source code under an open license, allowing researchers around the world to develop the technology. This move could accelerate the system’s impact in various scientific fields.

The development of MatterGen is a major part of Microsoft AI for Science The goal, which aims to accelerate the discovery of science using AI. This project is related to Microsoft’s Azure Quantum Elements platformwhich makes technology available to businesses and researchers through cloud computing services.

However, experts warn that while MatterGen represents a significant advance, the path from artificially produced products to practical products still needs extensive testing and refinement. The mechanical predictions, although promising, require experimental validation before industrial deployment.

However, technology represents a major area for using AI to accelerate scientific discovery. As Daniel Zügner, the project’s principal investigator, said, “We are very committed to research that can have a positive, real-world impact, and this is just the beginning.”