Google DeepMind has leveraged artificial intelligence (AI) to forecast the composition of over 2 million novel substances, a milestone that could soon enhance practical technologies.
In a recent publication in the scientific journal Nature, the AI company owned by Alphabet (GOOGL.O) revealed that nearly 400,000 of its conceptual material designs are on the verge of being actualized in laboratory settings.
The implications of this advancement span various sectors, including the enhancement of batteries, solar panels, and computer chips.
The exploration and creation of novel materials often entail significant expenses and time investments. For instance, the commercial availability of lithium-ion batteries, now ubiquitous in devices ranging from smartphones and laptops to electric vehicles, required approximately two decades of research.
Ekin Dogus Cubuk, a research scientist at DeepMind, expressed optimism about the potential of substantial enhancements in experimentation, autonomous synthesis, and machine learning models to drastically reduce the 10 to 20-year timeline typically associated with material development.
DeepMind’s AI was trained using data from the Materials Project, an international research consortium established at the Lawrence Berkeley National Laboratory in 2011, which collated information on approximately 50,000 known materials.
The company has announced its intention to disseminate its findings within the research community, aiming to expedite further advancements in material exploration.
Kristin Persson, the director of the Materials Project, highlighted the industry’s cautious approach to cost escalations and the time required for new materials to become economically viable.
DeepMind disclosed its plans to shift focus towards predicting the feasibility of synthesizing these novel materials in laboratory settings after successfully forecasting their stability using AI.
