The most important question of the future is not “What can robots do?” The real question is: What will humanity allow robots to do?

Once, robots were humanity’s greatest dream. They would do the heavy work for us, work tirelessly in factories, undertake dangerous tasks, and provide humans with more leisure time, prosperity, and security. Today, this dream is partially realized. Artificial intelligence writes, draws, codes, and analyzes; robots produce goods in factories. Autonomous systems are even taking on an increasingly important role on the battlefield. But behind this development, an increasingly pressing question emerges: If machines produce, algorithms make decisions, and robots fight, what will humanity’s economic, social, and moral place be?

This question is no longer a subject for science fiction literature. “Lights-out factories,” or “dark factories,” as they are called, are unmanned or nearly unmanned production facilities that can operate without human intervention, even without the need for light. According to Siemens, these types of factories are facilities with a level of automation that allows them to operate with near-zero on-site human intervention, even in the dark (https://www.siemens.com/en-us/technology/lights-out-factory).

This image is technologically impressive, but also symbolic. The factory is still producing, the machines are still running, goods are on the market, but the worker is no longer there.

Artificial intelligence and robotization are often described in terms of efficiency, innovation, and competitiveness. If a machine can do the same job faster, cheaper, and more accurately, the company wants to replace the worker with the machine. When a competitor does this, others are forced to follow suit to avoid falling behind…

A study published in March 2026, titled “The AI Layoff Trap” (https://arxiv.org/pdf/2603.20617), argues that this process could create a more dangerous economic trap than it appears. The article’s central argument is this: Even if companies act rationally individually, they can still enter an automation race that harms the system as a whole. Because workers are not only a cost of production but also consumers. When workers’ incomes decrease, the customer base that buys the companies’ products also weakens.

In other words, companies can replace workers with artificial intelligence and robots to reduce costs in the short term. But when all companies do the same, society’s purchasing power decreases. People who become unemployed or whose incomes fall consume less. Thus, companies weaken their own customer base.

The International Labor Organization’s 2025 update indicates that the highest exposure to generative AI is seen particularly in office and administrative occupations, and that exposure is also increasing in digitized professional and technical occupations (https://www.ilo.org/publications/generative-ai-and-jobs-2025-update). Therefore, the issue is no longer just about the factory worker, but also about the white-collar worker, the young graduate, the academic, the translator, the lawyer, the teacher, and even the doctor.

In such a scenario, it is not surprising that the rich get richer. Because the ownership of robots, artificial intelligence systems, data infrastructures, and platforms is mostly in the hands of capital owners. If labor incomes weaken while capital incomes increase, income distribution worsens further. This is not only an economic injustice, but also a risk of democratic breakdown. Because when economic power is concentrated, political influence is also concentrated.

Therefore, the debate on robotization is inevitably linked to the debate on the welfare state. If an ever-larger segment of society is excluded from regular, full-time, secure jobs, how will traditional social security systems survive? How will insurance systems built on work function when work itself is fragmented?

The article “The AI Layoff Trap” issues an important warning. According to the authors, a universal basic income can support people’s purchasing power, but it does not directly eliminate the fundamental incentive for firms to replace workers with artificial intelligence. So, basic income can be a social buffer, but it may not stop the automation race itself.

The article, therefore, discusses the idea of a “Pigouvian automation tax” as a more targeted tool that reflects the social cost of automation on firms. How this proposal would be implemented in practice is, of course, debatable. Which technology is a productivity boost, and which is excessive automation creating social costs? Will every firm using robots be taxed? Will small businesses and tech giants be evaluated the same way? These are not easy questions.

But the debate itself is important. Because the issue is not about stopping technology; it’s about democratically managing the social consequences of technology. Which jobs we will automate, which areas we will protect human labor in, and who will share the profits from increased productivity are all political choices.

On the other hand, it was assumed that activities such as scientific thinking, critical analysis, writing, teaching, and interpretation could not be easily automated. Today, this assumption is being shaken. Artificial intelligence systems can search literature, summarize texts, write statistical code, produce evaluations similar to peer-reviewed reports, and prepare course materials.

If universities and research institutions use artificial intelligence not to strengthen academic freedom, critical thinking, and scientific quality, but to reduce costs, standardize courses, reduce staff, and increase productivity pressure, then academia will also suffer from the social crisis of robotization.

The darkest dimension of robotization emerges on the battlefield. When technology, which replaces human labor in the factory, begins to decide about human lives on the battlefield, the issue is no longer just economics, but human rights.

Autonomous weapon systems are discussed as systems capable of selecting and striking specific targets without human intervention. At the United Nations level, “lethal autonomous weapon systems” have been on the agenda for years. The UN Office for Disarmament reports that Secretary-General António Guterres finds such systems “politically unacceptable” and “morally reprehensible,” and has called for their prohibition under international law (https://disarmament.unoda.org/en/our-work/emerging-challenges/lethal-autonomous-weapon-systems).

War is already one of the gravest moral crises in human history. If the decision to kill is increasingly left to machines, the threshold for war may also decrease. Because states whose soldiers face a reduced risk of death may more readily resort to military intervention. Robot soldiers, by making war appear “cleaner,” could actually make it more widespread.

According to a Reuters report from March 2026, international talks in Geneva emphasized the urgent need to advance rules on lethal autonomous weapon systems. It was noted that 128 states were considering a non-binding text, but binding global standards were not yet in place. This delay is significant because technology is advancing faster than law (https://www.reuters.com/world/progress-rules-lethal-autonomous-weapons-urgently-needed-says-chair-geneva-talks-2026-03-03).

If artificial intelligence and robotization are seen only as tools for cost reduction, workforce reduction, profit increase, and military superiority, the result could be a more unequal, insecure, and dangerous world. But if technology is linked to democratic oversight, social justice, ethical limits, and human rights, it can become a tool that facilitates human life rather than destroys human labor.

Robotization is inevitable, but what kind of robotization we will experience is not. This is a choice that is as much moral and political as it is technical. Humanity faces two paths: The first path involves robots producing, artificial intelligence making decisions, capital becoming even more concentrated, the middle class shrinking, people becoming insecure, and machines killing on battlefields. The second path involves using technology to improve human lives rather than devaluing human labor, sharing profits with society, strengthening the welfare state, maintaining academia and education’s humane character, and ensuring that decisions about life and death in war are not left to algorithms.

Therefore, the most important question of the future is not “What can robots do?” The real question is: What will humanity allow robots to do?