The man behind Google's groundbreaking Gemini model is taking on an even more ambitious endeavor: utilizing AI to potentially 'solve' diseases. Demis Hassabis, a child prodigy from North London, has always been captivated by the vastness of the universe, particularly the constellation Orion. This fascination with finding order in randomness led him to become one of the most influential figures in AI research and entrepreneurship.
Hassabis co-founded DeepMind, which was acquired by Google in 2014. DeepMind's AlphaGo made history by defeating the world's top player in the complex game of Go, marking a significant AI milestone. But it was AlphaFold 2, an AI system that predicts protein structures, that brought Hassabis and his colleague John Jumper the Nobel Prize in Chemistry in 2024. This achievement holds immense potential for understanding and treating diseases caused by protein misfolding, such as Parkinson's, muscular dystrophy, and certain cancers.
Hassabis's latest venture, Isomorphic Labs, aims to revolutionize drug discovery. The company's ambitious goal is to 'solve all disease' by creating breakthrough medicines for 'undruggable' conditions. This approach focuses on using AI to predict drug-target interactions at the molecular level, streamlining the pre-clinical trial process and making the idea of 'solving' diseases a tangible possibility.
Isomorphic's journey has been challenging, with the company initially raising funds from Google's parent company, Alphabet, and later securing $600 million in Series A funding. The startup is betting on developing precise, tech-driven drug design processes that will eventually become standard, akin to the design of airplanes. However, the odds are stacked against them, with the number of possible chemical compounds far exceeding the stars in the observable universe.
Drug discovery has traditionally been a hit-or-miss process, with a 90% failure rate in clinical trials. Isomorphic's approach aims to reduce this failure rate and accelerate the discovery and trial process by 50%. Hassabis's philosophy is that 'solving disease' is about creating a systematic, scalable process for discovering and designing drugs as needs arise, rather than eliminating illness entirely. This includes the potential for personalized medicine, where individuals could have their specific diseases phenotyped for tailored treatments.
The success of Isomorphic Labs could mark a turning point in drug discovery, but it also raises questions about the limits of AI in medicine. Will AI-driven drug design live up to its promise, or will it face challenges that even the most advanced technology can't overcome? The journey of Isomorphic Labs is one to watch, as it attempts to navigate the vastness of biological possibilities and bring about a new era in healthcare.
