The venture capital world loves a good morality play. When Joshua Kushner recently criticized his peers for chasing AI hype without discipline, the industry nodded along approvingly. Self-awareness feels good. But here's what we should actually be watching: the gap between what VCs say they value and who they actually fund.
The uncomfortable truth is that venture capital has built itself around a perverse incentive structure. It rewards pedigree, networks, and narrative over execution and judgment. And the people who benefit most from this arrangement have every reason to keep it exactly as it is.
Consider how VC actually works. Venture capitalists need to deploy large pools of capital into a portfolio of bets. They win by hitting outliers, not by being right most of the time. This creates a powerful incentive: invest in the kinds of founders who are most likely to generate story-worthy outcomes, whether those outcomes are real or imagined.
What kinds of founders are those? Founders with existing platforms. Founders from prestigious schools. Founders who have already succeeded before. Founders with connections to other wealthy people. Founders, frankly, who look and sound like previous successful founders.
This isn't a conspiracy. It's incentive architecture. A partner who backs a founder from Stanford with an impressive pedigree can tell their limited partners a compelling story, even if the company fails. The partner can explain that they were disciplined, that they backed quality, that the macro environment shifted. A partner who backs an equally talented founder from a non-prestigious background with the same outcome? That story is much harder to tell.
Over time, this shapes which founders get funded, which founding teams can scale, and which ideas get resourced. It's not that less connected founders build worse companies. It's that they get fewer chances to prove it.
The real kicker is that this system has become self-justifying. Because VCs have historically funded founders with certain advantages, those founders have become the cohort most likely to succeed statistically. Then VCs point to that track record and say: "See? We were right to fund people like that." The feedback loop closes. The prestige becomes the predictor, not the proxy for actual capability.
This matters because it narrows the pool of founders solving the world's problems. It biases capital toward certain geographies, certain demographics, certain school networks. It means that some of the most talented people who could build transformative companies never get a meaningful shot at venture funding.
The industry's recent hand-wringing about AI bubble dynamics is illustrative. Plenty of VCs are now admitting they may have chased euphoria. But notice who is making those admissions: established, well-connected people with reputational capital to spare and alternative ways to access deal flow. For founders without that same social infrastructure, a misjudgment isn't a learning moment. It's career death.
What would change this? Real accountability for capital deployment decisions would help. Not moral accountability, but actual consequences tied to judgment and returns, not prestige. Systematic effort to source founders from non-traditional backgrounds. Recognition that a founder's network size at day one is not the same as their ability to build something valuable.
But those changes would require VCs to admit that their current system serves their own interests quite well, regardless of how it affects founders or innovation broadly. That's a harder admission than nodding along when someone else critiques AI hype.
The venture capital industry isn't broken. It's working exactly as designed. The problem is that it was designed to serve venture capitalists and their limited partners, not necessarily to find and fund the best founders or the most important problems.
Until readers and founders understand whose incentives are actually being served here, that design will remain unchanged.