“We’re not doing 30 bets a year”: Vijay Pande on betting small after running $4 billion at a16z
Vijay Pande, who built Andreessen Horowitz’s life-sciences investment practice into one managing nearly $4 billion, has launched a smaller venture firm, VZVC, with investor Zach Werner. The firm plans to make only a handful of concentrated investments each year, operate without associates and use AI heavily, reflecting Pande’s view that focused backing is more effective than making dozens of bets.
Pande argues that AI could help turn biology from a largely discovery-led field into one that can be engineered, improving drug-target selection, development and eventually clinical trials. However, trials can still cost hundreds of millions of dollars, and only about 20% of drugs progress successfully from the first to the third trial; he says better AI models may eventually outperform animal testing, whose ability to predict human outcomes is limited.
- Pande has left a16z to build a smaller, concentrated AI-enabled investment firm.
- VZVC will make few investments rather than dozens each year.
- AI may improve drug development, but clinical trials remain costly and risky.
Both sides, in good faith
The strongest fair case each way — we don't pick a winner.
The case for
A small firm making a few concentrated investments could give each company more focused attention and let experienced investors make decisions without layers of process. Pande’s view is that AI may make biological research more predictable and engineerable, helping investors choose drug targets and develop treatments more effectively; over time, better models could also offer a useful alternative to animal testing.
The case against
Concentrating a fund in a handful of companies can magnify the consequences of any mistaken judgement, while a broader portfolio spreads risk across uncertain ventures. Biology and clinical development remain difficult to predict, trials are costly and few drugs advance successfully, so AI’s promise may not yet justify betting heavily on the idea that discovery can be made reliably engineerable.