AI can search through possibilities faster than any person. But this paper argues that the most important business opportunities are the ones nobody knew to search for—and that finding them depends on being a human being in the world. The debate behind this paper asks whether an economy could be planned centrally with data and prediction, and whether such a plan could recreate entrepreneurship.

Its central concern is that a centrally planned economy, even if powered by AI, may still be unable to recreate entrepreneurship. The paper identifies entrepreneurial discovery as the market function that current artificial intelligence cannot replicate. Human wants keep shifting unpredictably, so satisfying everyone is a moving target.

In a market economy, alert entrepreneurs notice unmet demands, think creatively, and use available resources in a new way to bring about innovation. That alertness means finding a mismatch between what exists and what people may want next.

The paper argues that this creative alertness requires sensory and emotional connection to a situation, and therefore a body. The argument turns on lived experience. A person who is materially involved in the world can notice possibilities in a market that would not be visible from information alone.

This practical know-how cannot be fully expressed as a set of statements, so a system without that lived involvement cannot exercise it in the same way. Such a system may optimize how existing resources are arranged, but it cannot deliver opportunities that were not already present in its training information.

Here is the key distinction: searching means looking through possibilities that have already been defined, while discovery means noticing something that was not being looked for. An opportunity can appear unexpectedly: discovery notices something that was not being looked for, before any deliberate search for it has begun.

The system discussed here explores combinations of ideas that a human mentor has already marked as relevant or promising. That means the human mentor still performs the entrepreneurial act of identifying which opportunities are worth pursuing. The operator also sets the situation’s possible outcomes, search costs, and trade-offs, so the system searches inside a human-defined frame.

Language models can lower the cost of searching, shorten the time needed to survey market conditions, and help brainstorm opportunities that have already been identified. But faster search is not discovery. It speeds up the use of recognized opportunities while leaving the prior act of recognizing the opportunity to a human agent shaped by institutions and history.

The argument does not require saying that human-level artificial intelligence is impossible in principle. Even if it appears, the paper’s position is that it would still be a form of cognition without lived bodily involvement. The stronger commitment concerns an artificial agent whose sensing, movement, and robotic parts are integrated closely enough to gather rich information about its surroundings.

The paper does not claim that such human-level bodily integration is likely, and it keeps open the earlier objections to artificial intelligence having genuine embodiment. The commentary presents the paper as an important angle on the connection between entrepreneurship and the market order.

Its conclusion is that entrepreneurship matters because it fosters innovation and the economy’s continuing change. For people outside economics, the practical implication is direct: AI may help organize known options, but new possibilities for care, work, and everyday life still depend on people noticing what no system was asked to find.

The paper’s central message is that AI may improve the search for known opportunities, but human entrepreneurs notice unexpected possibilities through lived experience. That difference shapes how innovation reaches patients, customers, and communities.