The current oligopolistic 'Cournot' state of AI labs will eventually shift to 'Bertrand' competition, where labs compete more on price. This happens once the frontier commoditizes and models become 'good enough,' leading to a market structure similar to today's cloud providers like AWS and GCP.
Related Insights
Firms like OpenAI and Meta claim a compute shortage while also exploring selling compute capacity. This isn't a contradiction but a strategic evolution. They are buying all available supply to secure their own needs and then arbitraging the excess, effectively becoming smaller-scale cloud providers for AI.
Early tech giants like Google and AWS built monopolies because their potential wasn't widely understood, allowing them to grow without intense competition. In contrast, because everyone knows AI will be massive, the resulting competition and capital influx make it difficult for any single player to establish a monopoly.
While US firms lead in cutting-edge AI, the impressive quality of open-source models from China is compressing the market. As these free models improve, more tasks become "good enough" for open source, creating significant pricing pressure on premium, closed-source foundation models from companies like OpenAI and Google.
The assumption that enterprise API spending on AI models creates a strong moat is flawed. In reality, businesses can and will easily switch between providers like OpenAI, Google, and Anthropic. This makes the market a commodity battleground where cost and on-par performance, not loyalty, will determine the winners.
Top-tier coding models from Google, OpenAI, and Anthropic are functionally equivalent and similarly priced. This commoditization means the real competition is not on model performance, but on building a sticky product ecosystem (like Claude Code) that creates user lock-in through a familiar workflow and environment.
Fears of a single AI company achieving runaway dominance are proving unfounded, as the number of frontier models has tripled in a year. Newcomers can use techniques like synthetic data generation to effectively "drink the milkshake" of incumbents, reverse-engineering their intelligence at lower costs.
Unlike traditional SaaS where high switching costs prevent price wars, the AI market faces a unique threat. The portability of prompts and reliance on interchangeable models could enable rapid commoditization. A price war could be "terrifying" and "brutal" for the entire ecosystem, posing a significant downside risk.
Major AI labs operate as an oligopoly, competing on the quantity of supply (compute, GPUs) rather than price. This dynamic, known as a Cournot equilibrium, keeps costs for frontier model access high as labs strategically predict and counter each other's investments.
The AI value chain flows from hardware (NVIDIA) to apps, with LLM providers currently capturing most of the margin. The long-term viability of app-layer businesses depends on a competitive model layer. This competition drives down API costs, preventing model providers from having excessive pricing power and allowing apps to build sustainable businesses.
Contrary to the 'winner-takes-all' narrative, the rapid pace of innovation in AI is leading to a different outcome. As rival labs quickly match or exceed each other's model capabilities, the underlying Large Language Models (LLMs) risk becoming commodities, making it difficult for any single player to justify stratospheric valuations long-term.
