A provocative claim is making the rounds in tech circles: AI might be ending venture capital as we know it.
The argument isn’t that VC will disappear, but that the traditional playbook—find a startup with a moat, fund it, wait for the 10x return—is breaking down in the age of AI.
The Moat Problem
On The Pomp Podcast, investor Jordi Visser laid out the core issue:
The traditional VC model depends on companies building defensible moats. But when AI can replicate features in weeks, how do you build a moat?
The Bottleneck Opportunity
Visser sees the real opportunity not in AI applications, but in the infrastructure layer:
This aligns with what we’re seeing in public markets—NVIDIA, the hyperscalers, and data center REITs have been the big winners, not the AI application companies.
What This Means for Founders
If the VC thesis is shifting, what does that mean for founders?
- Speed over moats: Ship fast, iterate faster. Your moat is execution speed.
- Infrastructure plays: Building picks-and-shovels might be safer than building apps.
- Bootstrap is back: If AI makes building cheaper, maybe you don’t need VC at all.
The irony? The technology that VCs are most excited about might be the one that disrupts their own business model.
Want to explore more podcast conversations about AI and investing? Search our transcript database.