Google DeepMind has presented at Google I/O an improved version of its artificial intelligence model that predicts not only the structure of proteins but also the structure of “all molecules of life.”

The results of the new AlphaFold 3 model will help researchers in the fields of medicine, agriculture, materials science, and drug development to test potential discoveries.

Previous versions of AlphaFold only predicted the structure of proteins. AlphaFold 3 goes beyond this and can model DNA, RNA, and smaller molecules called ligands, which expands the model’s capabilities for scientific use.

DeepMind says the new model shows a 50 percent improvement in prediction accuracy over previous models.

“With AlphaFold 2, it was a big milestone moment in structural biology and has unlocked all kinds of amazing research,” DeepMind CEO Demis Hassabis said at the conference. “AlphaFold 3 is a step along the path in terms of using AI to understand and model biology.”

AlphaFold 3 has a library of molecular structures. Researchers enter a list of molecules they want to combine, and then AlphaFold 3 uses a diffusion method to create a 3D model of the new structure. Diffusion is the same type of AI system used by AI image generators such as Stable Diffusion to arrange photos.

The new AI should be available “later this year.”