Meg Rithmire argues the US is paying for the frontier while China pays for diffusion. Her sharper question is who owns the stack the world relies on next.

Harvard Business School's Meg Rithmire makes a point (in a recent Harvard Business School Working Knowledge Q&A) I keep hearing in the Gulf: the US and China are not running the same AI race.

The US is paying for the frontier. Bigger models, bigger data centers, more private credit pointed at the LLM layer.

China is paying for diffusion, putting AI into manufacturing, logistics, and robots. Its models are cheaper and "OK in intelligence." Most buyers do not need more. In her words, they will take the one that costs 14% as much to run their factory floor.

Her sharper question is who owns the stack the world relies on for the next generation. That is decided by price and deployment, not by a leaderboard.

I see this in two places. Gulf buyers already hedge across US and Chinese models. And in robotics, the winners will be the teams that ship into a real factory, farm, or warehouse, not the ones with the best demo.

She also asks a question I cannot shake: will the US capital in the model layer earn a return? I would add one for any founder: what does your product cost per task if the model underneath gets 85% cheaper?

This perspective was originally published on LinkedIn. Join the discussion, add your thoughts, and follow for regular updates.

View on LinkedIn