Why One of VC's Biggest Biotech Names Is Betting Smaller — and What It Means for AI in Business

Vijay Pande helped build one of the most formidable biotech investment portfolios in Silicon Valley history. Overseeing roughly $4 billion in assets at Andreessen Horowitz, he had the resources to make broad, speculative bets across the life sciences. Now he has walked away from all of that — by choice — to run VZVC, a far smaller, AI-native fund that he says will make fewer, more deliberate investments.

In a new interview with TechCrunch's Connie Loizos, Pande is candid about the shift: "We're not doing 30 bets a year." The reasoning behind that restraint turns out to be more instructive than it first appears — not just for investors, but for any business leader trying to understand where AI is actually delivering real change.

Biology Is Becoming an Engineering Discipline

The core thesis driving Pande's new fund is a significant one. For most of its history, biology has been a science of discovery — researchers observe natural systems, document what they find, and slowly build knowledge over decades. Pande argues that AI is changing the fundamental nature of that work. Biology, he contends, is transitioning from a discovery science into an engineering one.

That is not a minor semantic distinction. Engineering disciplines operate with repeatability, predictability, and design intent. You define an outcome and build toward it. Discovery sciences are inherently messier — you follow the evidence wherever it leads, on timelines you cannot control.

If Pande is right, the implications extend well beyond drug development. The same transition — from reactive discovery to proactive engineering — is happening across sectors wherever AI has access to sufficient data and processing power. Manufacturing, logistics, financial modeling, customer operations: the pattern is consistent. AI does not just help you find answers faster. It changes what kinds of questions you can ask in the first place.

The Open Data Argument That Every Business Leader Should Hear

Perhaps the most striking position Pande takes in the TechCrunch interview is on data strategy. While many large players in healthcare and biotech have moved aggressively to lock down proprietary datasets — treating data as a competitive moat — Pande argues this is exactly the wrong approach.

His view is that open, shared datasets are what will actually allow AI to transform medicine. Walled-off data slows the entire field down. Collaboration, not hoarding, is how the science advances.

This is a genuinely contrarian take in an era where "data is the new oil" has become a reflex response. And it deserves serious attention from business strategists.

The instinct to protect proprietary data is understandable. But in practice, organizations that contribute to and draw from shared data ecosystems — industry benchmarks, open APIs, federated research initiatives — often move faster and build more capable models than those working in isolation. The competitive advantage, Pande implies, comes not from owning the data but from knowing how to use it.

For SMBs in particular, this framing is liberating. You do not need a data warehouse the size of a Fortune 500 company to benefit from AI. You need good integrations, quality inputs, and the right tools to act on what the data tells you.

The Deliberate Bet: Quality Over Volume

There is a broader management lesson in Pande's shift from 30 investments a year to a much tighter portfolio. At scale, quantity becomes a strategy of its own — you diversify because you cannot predict which bets will pay off. When you shrink the portfolio, every decision carries more weight, and the quality of your analysis has to improve accordingly.

Business teams deploying AI tools face a similar tension. The temptation is to adopt as many tools as possible, treating AI as a volume play. The more automations you run, the more workflows you optimize, the more you assume the gains will compound. But without deliberate evaluation of where AI actually creates meaningful leverage, you end up with tool sprawl and marginal returns.

The better model is what Pande is demonstrating: fewer, better-chosen applications, deployed with clear intent and measurable outcomes. For operations, marketing, and customer-facing teams, that means identifying the two or three workflows where AI genuinely changes your output quality — not just your speed.

If you are working through that process right now, platforms like WRRK.ai are built specifically to help teams identify and operationalize those high-leverage AI applications without the overhead of enterprise-scale implementation.

Clinical Trials and the Limits of Hype

Pande is also refreshingly honest about where AI has not yet solved the hard problems. Clinical trials, he notes, remain brutally expensive. The regulatory, logistical, and human complexity of running trials does not disappear because you have a better model for target identification. AI accelerates the science; it does not yet replace the infrastructure.

That kind of grounded assessment is worth applying to your own operations. AI is a force multiplier, not a replacement for sound processes and experienced judgment. The businesses seeing real returns are the ones that treat it that way. For more on building that foundation, see our overview of practical AI tools for business teams.

Original reporting by Connie Loizos, TechCrunch AI. Published August 29, 2026. Read the full interview at TechCrunch.

Frequently Asked Questions

What is Vijay Pande's new venture fund VZVC focused on?

VZVC is an AI-native venture fund founded by Vijay Pande after he departed Andreessen Horowitz, where he managed approximately $4 billion in biotech investments. The new fund takes a more focused approach — fewer investments, selected with higher conviction — with a thesis centered on AI's ability to transform biology from a discovery science into an engineering discipline.

Why does Vijay Pande believe open data is better than proprietary data for AI in medicine?

Pande argues that shared, open datasets — rather than walled-off proprietary ones — are what will enable AI to meaningfully advance medicine. His position is that data hoarding slows scientific progress across the board, and that collaboration produces better AI outcomes than isolation. This runs counter to the dominant strategy of many large healthcare and technology players.

What can small and medium-sized businesses learn from AI investment trends in biotech?

The key lessons are focus and intent. Just as Pande is moving away from high-volume, speculative bets toward fewer, more deliberate investments, SMBs should resist the urge to adopt every available AI tool. The businesses generating real returns from AI are those that identify specific, high-value workflows and deploy AI with clear goals — rather than treating adoption as a numbers game.

Discover how WRRK.ai helps business teams deploy AI where it actually matters — without the complexity.

Why One of VC's Biggest Biotech Names Is Betting Smaller — and What It Means for AI in Business