Over the past couple of years, I have followed the work of DeepMind’s groundbreaking developments. I have been most fascinated by the intersection of biology and AI through AlphaFold, an AI program that performs predictions of protein structure. AlphaFold generates the most accurate 3D protein models of all time. With this level of accuracy, we can predict whether a protein will become an antibody, hormone, membrane protein, or any one of approximately 20,000 proteins in the human body. AlphaFold examines amino acid sequences, finding which sections have a high probability of mutation and correlation with the final 3D structure to determine protein structure. The prediction of protein folding helps determine the function of a protein and how it works. This means that we can predict how certain drugs can bind to proteins and their effects on human health. In the past, techniques such as nuclear magnetic resonance, cryo-electron microscopy, and X-ray crystallography would be used to determine protein structures, but they would require significant funding and years of trial and error to achieve what AlphaFold can do in minutes or hours. DeepMind open-sourced the AlphaFold code and a database of protein structures, which is comparable to Rosetta Stone because of its ability to translate human genetic code into specific proteins and their functions. DeepMind helped jumpstart the drug development for COVID-19 by predicting the structure of numerous proteins in the SARS-CoV-2 virus which was previously unknown. This predictive modeling was foundational to pandemic responses for COVID-19 and will continue to be useful in pandemic prevention in the future. Therefore, I believe AlphaFold is the most significant application of AI in advancing scientific knowledge today and for years to come.