CharlieKey takeaways from your favorite podcasts
← 20 October 2026The Cognitive Revolution

A World Model of Life: Biohub's Alex Rives on Virtual Cells, Measurement, & the $1.8B Data Push

Alex Rives argues that virtual cell models are limited by what we can measure, so Biohub is investing in new measurement technology and large data sets to build an information-theoretic model of the cell.

Watch on YouTube ↗

Charlie's summary

  1. Today's virtual cell models fail because biology's biggest limit is measurement, not just modeling

    Alex says progress needs three things: better modeling paradigms, data generated at scale, and tools that observe what matters. The transformer lacks the multi-scale structure a cell needs, and much of the cell “we just really can't measure.” So he doesn't think models called virtual cells today meet the definition of a model that could simulate the cell and serve as an oracle for experimental biology.

    ▶ 9:11
  2. Aim first for an information-theoretic model of the cell: a probabilistic transition function toward a world model of life

    Alex thinks a full world model of life, from atomic structure to tissue development, is far off. What AI can reach in about five years is an information-theoretic model that treats the cell as an information processor encoded by the genome. Its first piece is a transition function that maps a cell state and inputs to a distribution of next states, not a single deterministic one. Where data is scarce, inductive bias matters more.

    ▶ 14:18
  3. The $1.8B Virtual Biology Initiative is a data push on contextual maps of cells

    Biohub announced a partnership with the Department of Energy, NIH, Google DeepMind, Isomorphic and Meta to coordinate data generation. Priorities include spatial biology that shows cells in tissue context, and links between molecular measurements and imaging, at the scale of hundreds of billions of cells or more. Alex says the goal is to move faster and create the first data sets for training contextual models of the cell, and he invites other funders and scientists to join.

    ▶ 23:05
  4. New measurement technology drives paradigm shifts, as with the laser phase plate

    The laser phase plate adds phase contrast to electron microscopy by using a powerful standing-wave laser to shift the electron beam's phase. For cryo-electron tomography this makes the very noisy images clear enough to pick out individual proteins and macromolecules in their living context. Alex says biology differs from other AI domains mainly in how limited our ability to observe what matters is. Biohub therefore pushes more parameters per cell, more cells, time and molecular detail.

    ▶ 26:12

Give your email to unlock the rest of the key points — for every episode.

Do more with Charlie

Free — sign in with your email, no password.