a16zMark Zuckerberg & Priscilla Chan: How AI Will Cure All Disease
At a glance
WHAT IT’S REALLY ABOUT
CZI Biohub bets on AI tools to accelerate biology breakthroughs
- CZI’s core strategy is tool-building: create shared, open resources that let the broader scientific community move faster than traditional grant-by-grant funding allows.
- They argue biology lacks an organizing “periodic table”-like foundation, motivating efforts like cell atlases, standardization, and virtual cell models as new primitives for discovery.
- Biohub’s differentiator is pairing frontier biology with frontier AI in one organization, enabling a tight feedback loop where experiments produce data tailored to model gaps.
- “Virtual cells” are positioned as a model-organism-like platform that can de-risk hypotheses in silico, enabling more ambitious experiments despite the cost and slowness of wet labs.
- Success is measured less by CZI producing drugs and more by catalyzing an ecosystem—startups, academia, pharma—using Biohub tools to enable precision medicine across diseases that are increasingly treated as individualized/‘rare’ conditions.
IDEAS WORTH REMEMBERING
5 ideasThe fastest path to medical progress is often a new tool, not a single discovery.
They frame scientific leaps (microscope/telescope analogies) as tool-enabled, and position Biohub as an engine for building tools that let many labs debug biology faster.
Philanthropy can fund the “missing middle” of science: expensive, long-horizon infrastructure.
NIH-style grants optimize for smaller, near-term projects; Biohub targets 10–15 year, $100M–$1B efforts (datasets, instrumentation, models) that are hard to justify in traditional funding structures.
Standardization can create compounding “network effects” in science.
Cell by Gene began as an annotation bottleneck fix for single-cell data; by making annotation easy and standardized, it pulled in broader community contributions (they claim ~75% came externally), turning a tool into an ecosystem.
Virtual cells are meant to make biology more ‘testable’ earlier, enabling higher-risk ideas.
Because wet-lab work is slow and career incentives punish failure, a useful (even imperfect) simulation can provide directional signals to de-risk hypotheses before committing time and money.
The ‘virtual cell’ roadmap is explicitly hierarchical and modular.
They describe building from protein models (e.g., folding/structure) into cellular models and then to larger systems (e.g., virtual immune system), combining specialized models into more general “world models” over biology.
WORDS WORTH SAVING
5 quotesWhen we first set out that-- the goal to cure and prevent disease by the end of the century, people-- Like, honestly, most scientists couldn't look at us with a straight face.
— Priscilla Chan
And it was true because if you just decided to spend the money funding the next best grant for every single lab in the country, like, you-- there's no pathway to that being true.
— Priscilla Chan
It's kind of this crazy thing that we're, you know, here in, you know, 2025, and there's not the kind of periodic table of elements equivalent for biology.
— Mark Zuckerberg
And, and so those are rare, like and, and but I really think most diseases should be thought of as rare diseases because each one of our biology is different.
— Priscilla Chan
But if you had a virtual cell model where you could simulate really high quality biology, you could actually then start testing and tinkering on the computational side and like ask riskier questions, things that would've been expensive and ti- costly in terms of time and resources to do in the lab, and actually see if there is promise doing the experiments in silico before you make the time and money investment in the wet lab.
— Priscilla Chan
High quality AI-generated summary created from speaker-labeled transcript.
Get more out of YouTube videos.
High quality summaries for YouTube videos. Accurate transcripts to search & find moments. Powered by ChatGPT & Claude AI.
Add to Chrome