LLMs are becoming commoditized. Like gas from different stations, models can be swapped based on price or marginal performance. This means competitive advantage doesn't come from the model itself, but how you use it with proprietary data.

Despite being a commodity business with high costs and low defensibility, AI foundation models command massive valuations. They function as a 'hope' asset where investors park capital based on narrative, similar to how gold is used in uncertain times, rather than on financial fundamentals.

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.

Leading AI models are becoming increasingly similar in capability. This rapid convergence suggests the underlying technology is becoming a commodity, and competitive advantage will likely shift to user interface, distribution, and specific applications rather than the core model itself.

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.

Comparing AI to 1995-era internet bandwidth, the hosts argue that selling raw 'intelligence' is a low-margin, commodity business. The significant financial upside will be captured not by the infrastructure providers, but by the creators who build novel applications and experiences using that intelligence as a building block.

If AI makes intelligence cheap and universally available, its economic value may collapse. This theory suggests that selling raw AI models could become a low-margin, utility-like business. Profitability will depend on building moats through specialized applications or regulatory capture, not on selling base intelligence.

Unlike cable or power companies that benefit from regional monopolies, AI intelligence is a globally competitive, frictionless market. This dynamic is 'so much worse' for business because it allows for perfect arbitrage, driving the price of intelligence toward zero and making it incredibly difficult to build a sustainable, high-margin business on the infrastructure layer.

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.