Essays - Benedict Evans · Benedict Evans
articleinsightWays to think about token pricing
9 July 2026Strategy
Source excerpt
About Essays - Benedict EvansAI is in a supply crunch today, but what happens when we come out of it? How and where will supply, demand, price, capacity and capex get back into equilibrium? Today, model labs can name their price, but why won’t they end up as low-margin commodity infrastructure?
This is a limited feed-provided excerpt, not the full original work.
Apollo layer
What a founder can learn
Founder takeaway
Do not base your AI business on today’s supply constraints or model pricing. Plan for inference capacity to expand and foundation models to become cheaper and less differentiated; build advantage in proprietary workflows, customer relationships, data, or distribution instead.
Why it matters
If model access becomes commodity infrastructure, margins will migrate to businesses that differentiate above the model layer. Stress-testing this shift now helps you avoid fragile unit economics and invest in more durable sources of value.
Put it to work
Model your business under a 10× drop in token prices: what becomes more valuable, what loses pricing power, and which part of your advantage remains difficult to copy?