Priscilla Chan Sees AI Models of Cells as the Next Leap in Biology and Medicine

Finding the right medicine to treat conditions like anxiety and depression can be a frustrating and time-consuming process. Doctors often start patients on medications that are generally well-tolerated and effective for most people, but these drugs don’t work for everyone. Sometimes, they can even cause severe side effects. It can take months of trial and error to find a medication that works for a specific individual. Despite the challenges, there is hope that advancements in artificial intelligence (AI) could fundamentally change how we approach this problem.

At the South by Southwest (SXSW) conference in Austin, Texas, Dr. Priscilla Chan, co-founder of the Chan Zuckerberg Initiative and a pediatrician by training, spoke about the potential of AI to revolutionize biomedical research. Chan emphasized that if doctors could test drugs against generative AI models of individual patients’ cells and systems, the process of finding effective treatments could become much more efficient. She believes that AI could help answer some of the most pressing questions in biology and medicine, leading to breakthroughs that were previously unimaginable.

The potential of AI to transform science and medicine is vast, but it’s important to manage expectations. While AI could significantly speed up research, it won’t replace the hard work of scientists in labs. Instead, it will act as a tool to help researchers identify promising drug candidates and generate new hypotheses. For example, AI models could analyze vast amounts of biological data to predict how different compounds might interact with specific cells or proteins in the body. This could save researchers a tremendous amount of time and resources, as they would only need to test the most promising candidates in physical labs.

Dr. Chan also highlighted the vast gaps in our understanding of human biology. While we’ve made incredible progress in fields like genetics—such as mapping the human genome—there is still so much we don’t know about how the body works. Chan used the analogy of a Lego kit to illustrate this point. Just as the instruction manual for a Lego Millennium Falcon provides a roadmap but doesn’t show how every individual piece fits together, our understanding of genetics gives us a broad outline of how life works, but we’re still in the dark about many of the details. This lack of understanding is especially problematic when it comes to treating individual patients, as everyone’s biology is unique. AI could help bridge this gap by creating highly personalized models of individual patients, allowing doctors to tailor treatments to their specific needs.

One of the most exciting possibilities of AI in medicine is its ability to accelerate the development of new drugs and treatments. Current research methods are often slow and expensive, requiring scientists to test ideas in physical labs over months or even years. While these labs will always be essential, AI can help prioritize which ideas are worth pursuing in the first place. By analyzing existing data and identifying patterns that humans might miss, AI models can suggestion potential drug candidates that are more likely to work. This could significantly reduce the number of failed experiments and speed up the discovery process. Of course, AI isn’t perfect—it might suggest ideas that don’t work or aren’t feasible. That’s why human scientists will always play a critical role in evaluating and refining the ideas generated by AI.

Looking ahead, Dr. Chan sees AI as a potential game-changer for biomedical research, comparable to landmark inventions like the microscope, X-ray, or MRI. These technologies transformed our understanding of the human body and paved the way for countless medical advancements. AI could have a similar impact, enabling scientists to ask new kinds of questions and explore previously unimaginable possibilities. While it may take years or even decades for the full potential of AI in medicine to be realized, the hope is that it will lead to a future where treatments are more personalized, effective, and accessible to everyone. As Chan said, “Health and medicine, it moves in leaps. There are decades when research gets stuck, and then someone invents a new technology that completely changes how we see the human body.” With AI on the horizon, we may be on the cusp of one of those transformative leaps.