Traditional business school wisdom on startup value creation is incomplete. Here’s a typical critique:

“[The startup] does not fundamentally change the capitalization model of the enterprise, nor the unit economics of its products or services.”

This completely misses a key factor that derives competitive advantage: Customers don’t care about your unit economics – they care about perceived value: what they get versus what they give up.

Stripe charges the same 2.9% + 30¢ as PayPal; similar underlying economics but a 40% greater valuation. Why? Stripe turned a three-week payments integration into seven lines of code. They didn’t innovate on the business model – they demonstrated execution arbitrage.

Here’s the paradox: AI democratizes capability but concentrates returns. Every startup can deliver greater perceived value using APIs available to anyone with a credit card. None can defend it. In the AI era, competitive advantage now lasts exactly as long as it takes your competitor to write the same prompt.

So where’s the opportunity? In the gaps large players won’t touch – either because the TAM looks too small, the workflow too specialized, or the data too proprietary. The race is for startups to embed so deeply into these niche workflows that by the time they matter, they’ve become the de facto infrastructure.

Perhaps this is the end state: thousands of AI startups delivering incredible perceived value in narrow domains, all hoping to become important enough to acquire but not threatening enough to copy. We’re not building the next generation of tech giants – we’re building the features for the current ones.