# AI helps design new materials that work in the real world

DevFeed: [AI helps design new materials that work in the real world](<https://devfeed.tech/articles/ai-helps-design-new-materials-that-work-in-the-real-world-37941.md>)

Original publisher: [Read original article](<https://news.mit.edu/2026/ai-helps-design-new-materials-that-work-in-real-world-0826>)

Author: Zach Winn | MIT News

Published: 2026-08-26T09:00:00Z

Content type: news

Language: en

Sources: [MIT AI News](<https://devfeed.tech/sources/mit-ai-news.md>)

Topics: [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Framework](<https://devfeed.tech/topics/framework.md>), [Simulation and Design](<https://devfeed.tech/topics/simulation-and-design.md>), [Crystal](<https://devfeed.tech/topics/crystal.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [bowen-yu](<https://devfeed.tech/tags/bowen-yu.md>), [chemical-engineering](<https://devfeed.tech/tags/chemical-engineering.md>), [chemistry](<https://devfeed.tech/tags/chemistry.md>), [computer-chips](<https://devfeed.tech/tags/computer-chips.md>), [computer-science-and-technology](<https://devfeed.tech/tags/computer-science-and-technology.md>), [crystal](<https://devfeed.tech/tags/crystal.md>), [crysvcd](<https://devfeed.tech/tags/crysvcd.md>), [data-centers](<https://devfeed.tech/tags/data-centers.md>), [department-of-energy-doe](<https://devfeed.tech/tags/department-of-energy-doe.md>), [diffusion-models](<https://devfeed.tech/tags/diffusion-models.md>), [dmse](<https://devfeed.tech/tags/dmse.md>), [hao-tang](<https://devfeed.tech/tags/hao-tang.md>), [heather-kulik](<https://devfeed.tech/tags/heather-kulik.md>), [ju-li](<https://devfeed.tech/tags/ju-li.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [materials-design](<https://devfeed.tech/tags/materials-design.md>), [materials-discovery](<https://devfeed.tech/tags/materials-discovery.md>), [materials-science-and-engineering](<https://devfeed.tech/tags/materials-science-and-engineering.md>), [mingda-li](<https://devfeed.tech/tags/mingda-li.md>), [mouyang-cheng](<https://devfeed.tech/tags/mouyang-cheng.md>), [national-science-foundation-nsf](<https://devfeed.tech/tags/national-science-foundation-nsf.md>), [nuclear-science-and-engineering](<https://devfeed.tech/tags/nuclear-science-and-engineering.md>), [paper](<https://devfeed.tech/tags/paper.md>), [physics](<https://devfeed.tech/tags/physics.md>), [research](<https://devfeed.tech/tags/research.md>), [school-of-engineering](<https://devfeed.tech/tags/school-of-engineering.md>), [school-of-science](<https://devfeed.tech/tags/school-of-science.md>), [semiconductors](<https://devfeed.tech/tags/semiconductors.md>), [weiliang-luo](<https://devfeed.tech/tags/weiliang-luo.md>), [weiwei-xie](<https://devfeed.tech/tags/weiwei-xie.md>), [yongqiang-cheng](<https://devfeed.tech/tags/yongqiang-cheng.md>)

## AI overview

MIT researchers developed CrysVCD, a framework that applies chemistry-based valence constraints before material generation to improve the stability of generated designs. In tests, it achieved high lattice-dynamics stability in nearly 70 percent of computational material generations and supported targeting properties such as high thermal conductivity and high dielectric constant.

## Source excerpt

The "CrysVCD" tool developed at MIT could cut the huge amounts of time and money spent on screening out chemically unstable designs.