TBPN The Cournot Equation, Micron’s $200B Bet, Hollywood vs. Seedance 2.0 | Diet TBPN
42 snips
Feb 18, 2026 They unpack Cournot-style competition and how oligopolies shape AI labs' supply choices. The conversation jumps to inference economics, margins, and who pays for frontier access. Big hardware moves get attention with Micron’s massive memory gamble. Hollywood drama and synthetic likeness disputes surface alongside open-source model dynamics and safety concerns.
AI Snips
Chapters
Books
Transcript
Episode notes
AI Labs Compete On Supply, Not Price
- When a market has few dominant AI labs, they compete by choosing supply rather than price, creating a Cournot-like oligopoly.
- This supply competition drives races for GPUs, data centers, and low-latency capacity that shape frontier access and pricing.
Inference Is A High-Margin Manufacturing Business
- Inference behaves like manufacturing with positive contribution margins driven by GPUs, power, and engineering overhead.
- Labs can have healthy inference margins even while overall company profits are negative due to expensive model training.
Profitable Models, Unprofitable Firms
- Companies may report per-model gross profits while the firm loses money because training the next generation scales exponentially.
- The Cournot equilibrium emerges when training scale-up stabilizes and supply decisions balance R&D and marginal serving costs.
