TheMarketForIdeas

Robots have become an active part of our economic life. They are already present on the stock market where they act almost independently, buying and selling assets and resources although they are still following the requirements previously given by an entrepreneur who shoulders the risk for its actions. The development of sufficiently complex models that are able to predict and alter the conditions under which they operate will lead us, at some point, to question the role of man in the whole economic mechanism.

In terms of efficiency, although today man can still compete with algorithms, it is safe to assume that in the future this competitiveness will become a thing of the past. Perhaps, like computers in chess, a human against a computer (or advanced algorithm) may not stand a chance, but a human (or entrepreneur) together with a computer (a so-called “centaur”) may surpass what we thought of as the limits of the production of goods and services.

Today there exists a significantly pessimistic sentiment regarding the prospects that these advanced tools – ranging from physical robots to virtual bots and various types of artificial intelligences – can bring to humanity. Artificial intelligence is seen as a risk that requires either a brake on technological development or regulation within a policy framework rather than a tool for economic development.

The unintended consequences of such regulations remain unclear, particularly when they arise from a deep misunderstanding of the mechanisms that make these tools possible and their function within the market process.

Like any other human invention applied to production, it cannot be argued that robots and artificial intelligences represent a complete paradigm shift, nor that they will replace the production structure in any fundamental way.

AI has rekindled discussions about its potential role in shaping both human life and economic organization. Some now hope that the market economy will disappear as a consequence of mass unemployment caused by automation or through a completely novel type of organizing the economy.

These hopes are unfortunately unjustified. Although the technological advances are admirable, the problem of managing an economy is not an engineering problem involving static calculations that can be solved if all the variables and data are properly formulated. In other words, the problem is not a computational issue because the information needed to centrally plan such an economy requires a different type of epistemic knowledge that is distinct from the one used in robots or AI agents that we currently have access to.

The underlying epistemic problem involves local and tacit knowledge (Oğuz, 2010) that is based on the subjective preferences of the market participants, which are continuously changing through time and after each consumption cycle.

Boettke and Candela (2023), for instance, reconceptualize the problem of economic calculation in a centralized economy using AI as a central planner. A lack of economic exchanges for the factors of production will necessarily lead to an absence of economic prices, making the economic calculation impossible.

Of course, these highly advanced technological tools could make predictions quite specific to consumer goods, having already decades of development to be able to predict with high accuracy what the consumer wants to “consume”. These algorithms have been optimized to deliver specific content, sometimes being able to make predictions by spotting patterns that not even the user knew about themselves. The problem would only be partially solved, however, because the problem of economic calculation as originally postulated by Mises (2008) is relevant to the prices of the means of production and not to that of consumer goods. Predicting consumer preferences and behavior is a completely separate problem from the problem of calculating how to efficiently use capital goods. The former could be learned from big data while the latter requires prices and entrepreneurial judgement.

Without much reviewing of the technical process of how these robots will work, there will certainly be problems with their productive capabilities. In other words, there will be limited processes and calculations that they will be able to do. Evaluating the potential commercial and industrial benefits will inevitably involve trade-offs.

Within the market, i.e. the institutional arrangement where they can operate most effectively, companies offering robot-generated goods and services will be constrained by the very structure of the economy not to eliminate their consumers (Mises, 2018). In the absence of consumers, who have been totally replaced by machines, the question arises as to who will be the ones to buy the products. Moreover, production is not an autotelic phenomenon. It is not performed just for the sake of production, but in order to eventually lead to consumption. The crucial point is that, ultimately, technology must serve consumption and reflect the market preferences.

It is our desire to consume that will ultimately drive the output generated by these new tools, and the more complex our desires, the higher the opportunity cost that these new production methods will have. Only in a relatively free economy will we be able to assess whether it is more efficient to use these tools to make calculations about the precise workings of a nuclear reactor, or whether it is more efficient to use generative capabilities to make Ghibli-style pictures (Pendey, 2025).

Outside the market, we will of course be able to say a few things about human nature and the dominant values in society, especially when these are accompanied by real consequences that will be seen predominantly in the material world.

The labor factor is unlikely to become abundant even with the addition of new labor-saving instruments. Human desires are infinite and will continue to change even if the economic landscape could be more abundant in absolute terms of goods and resources that can be consumed. A possible reduction in the scarcity of labor will not mean that we will get rid of the resource scarcity problem, but it could positively contribute to a reduction in the undesirable aspects of production limited to a finite number of workers.

It would be naive of us to think that these technologies are free of problems. They can aggravate or raise new ones. But it is important to understand both their disruptive potential and their benefits. A balanced approach should highlight in particular the limitations of the applications of such technologies. Part of the interest in the subject has been due to an initial astonishment at the novelty of large language models (LLM) like ChatGPT.

Markets are robust and constantly adapting. They initially integrated scripts and pre-programmed robots and they will also integrate more advanced tools for more complex processes if this proves desirable from the consumers' point of view and if this process is economically feasible.

At the moment, a greater danger than technologies and robots are our disproportionate reactions to this relatively new and incomprehensible phenomenon. Particularly because they are not yet fully understood, we should remain skeptical of initiatives that seek to completely reorganize society around them.

References:

Boettke, P.J. and Candela, R.A. (2023) ‘On the feasibility of technosocialism’, Journal of Economic Behavior & Organization, January, pp. 44–54. Available at: https://doi.org/10.1016/j.jebo.2022.10.046.

Mises, L. von (2008) Economic calculation in the socialist commonwealth. Auburn, Ala: Ludwig von Mises Institute, Auburn University.

Mises, L. von (2018) Acțiunea umană: Tratat de teorie economică. București: Editura Institutului Ludwig von Mises România.

Oğuz, F. (2010) ‘Hayek on tacit knowledge’, Journal of Institutional Economics, 6(2), pp. 145–165. Available at: https://doi.org/10.1017/s1744137409990312.

Pandey, N. (2025) The ChatGPT Studio Ghibli trend: hurtling towards the point of no return. The Boar, 12 April. Available at: https://theboar.org/2025/04/the-chatgpt-studio-ghibli-trend-hurtling-towards-the-point-of-no-return/.

Photo source: PxHere.com.