AINo Priors
Mark Zuckerberg, Priscilla Chan, and Alex Rives discuss Biohub's mission to accelerate scientific progress through open-source tools and frontier AI combined with frontier biology. They delve into breakthroughs like ESM fold for protein design, the advantages of their non-profit model for broad impact, and their vision for personalized medicine by understanding biological systems hierarchically.
- Biohub began with an audacious goal from Mark Zuckerberg: to "cure, prevent and manage all disease by the end of the century," a vision Priscilla Chan admits was met with laughter by Nobel-winning scientists. Read →
- For breakthroughs in complex fields like frontier biology, a non-profit structure can achieve "bigger impact" by getting tools into more scientists' hands faster than a for-profit model. Read →
AILatent Space
This episode features Alex Rives from Biohub discussing the application of the 'bitter lesson' (scaling laws and empirical evidence) to protein biology through the ESMC language model. Rives explains how massive datasets, particularly metagenomics, enabled ESMC to build a comprehensive 'world model' of proteins, facilitating the design of novel therapeutics like antibodies and offering deep insights into protein function through mechanistic interpretability. The discussion also covers Biohub's ambitious Virtual Biology Initiative, aiming to integrate advanced experimental techniques with AI to create predictive digital representations of biological systems, from molecules to cells, to accelerate scientific discovery and combat disease.
- The "Bitter Lesson" applies to biology: Alex Rives, a key figure at Biohub, champions the principle that empirical scaling with massive data often outperforms human intuition or handcrafted rules in building powerful AI, even for complex protein biology. Read →
- ESMC's Blind Insight: The ESMC protein language model, trained only to predict amino acids from sequences, developed a "world model" of proteins so accurate it intrinsically understood complex biological concepts like the 'nucleophilic elbow'—without any prior biological input. Read →