Science · Technology and Human Behaviour

When Machines Challenge the Meaning of Being Human

Artificial intelligence is changing more than the tasks people perform. Research suggests that fear about the future of an entire profession can affect how workers understand their place, their organizations and their responsibilities.

For thousands of years, human beings have measured progress by the things they have learned to do. We shaped stones, cultivated fields, built cities, calculated the movements of planets and eventually constructed machines capable of exploring the physical universe. Each new tool extended the boundaries of what a human being could accomplish. For much of this history, machines amplified our muscles. Engines could lift what our bodies could not, and computers could calculate what our minds would take years to process. Artificial intelligence introduces a different possibility: a machine that can increasingly perform tasks we once considered evidence of human intelligence itself. It can write, translate, analyse, design, recognise patterns and participate in decisions. This raises a question that goes beyond economics or employment. What happens to a species when the tools it creates begin to challenge the abilities through which it understands its own value?

A research paper published in Occupational Health Science in September 2026, titled “In the Shadow of AI: How Industry Exposure to AI and Occupation Insecurity Fuel Moral Disengagement and Counterproductive Work Behaviors”, explores an unexpected consequence of this transformation. Rather than examining only how many jobs artificial intelligence might replace, the researchers investigated how the perceived threat of technological replacement could influence the psychology and behaviour of workers. Their central concern is occupation insecurity: the fear that technological advancement may threaten not simply a person's current position, but the future viability of the entire profession. There is a substantial psychological difference between losing a job and beginning to believe that the knowledge accumulated over an entire career may no longer have the same value. The first threatens a source of income. The second can threaten a person's understanding of who they are.

Imagine someone who has spent twenty years learning to interpret financial information, analyse complex documents or develop technical solutions. Their knowledge is not merely a collection of skills; it is the result of thousands of hours of observation, practice, mistakes and accumulated experience. Their profession has become part of their identity. Now imagine that an artificial intelligence system begins performing significant portions of that work in seconds. The machine does not need to understand what the work means to the person who performs it. It only needs to produce an output that an organization considers useful. The employee is suddenly confronted with a difficult comparison between biological experience accumulated over decades and computational capability that can be replicated across millions of operations. Even when the machine cannot replace the complete profession, the perception that it might eventually do so can create uncertainty about the future.

The researchers used three waves of survey data from workers across eighteen industry sectors with varying levels of exposure to artificial intelligence. They found that occupation insecurity was positively associated with moral disengagement, which in turn was positively associated with counterproductive workplace behaviours. The indirect relationship became stronger as industry-level exposure to AI increased. Moral disengagement describes a psychological process through which people rationalise actions that would ordinarily conflict with their own ethical standards. Counterproductive workplace behaviours are actions that can harm an organization or the people working within it. The study does not establish that AI directly causes workers to behave unethically, but it identifies a relationship between anxiety about the future of a profession and how some employees justify and report their conduct at work.

To understand why this matters, consider how people respond when they believe an institution no longer values their contribution. An employee who feels increasingly threatened by technological change may begin to interpret organizational expectations differently. They might feel that their loyalty is not being reciprocated, that management is preparing to replace them without recognising the years they have invested, or that the organization's demands are no longer fair. Under these circumstances, people can construct explanations that make previously unacceptable behaviour appear reasonable. They may withdraw cooperation, neglect responsibilities or justify cutting corners. This does not mean that every threatened worker becomes dishonest or deliberately destructive. It means that insecurity can change the way people rationalise their decisions, especially when they perceive the transformation as something happening to them rather than something in which they have a meaningful role.

There is an irony here. Companies often introduce artificial intelligence to increase efficiency, reduce costs and improve productivity. Yet an organization is not simply a collection of tasks that can be transferred from people to algorithms. It is a social system sustained by cooperation, trust, communication and shared responsibility. An automated system may process thousands of documents efficiently, but the larger organization still depends on people who understand its objectives, recognise exceptions and take responsibility for consequences. If the introduction of AI creates widespread uncertainty, employees may become less willing to share knowledge, collaborate or contribute beyond their immediate responsibilities. An organization can therefore become technologically more capable while its human relationships become less effective. The machine may be functioning exactly as designed, while the social system surrounding it begins to weaken.

This is one of the problems in how we measure technological progress. We frequently evaluate artificial intelligence through benchmarks, processing speed, accuracy and economic productivity. These measurements tell us something about what machines can do, but much less about what happens to the people whose lives are reorganised around those capabilities. A profession is not merely a collection of repeatable operations. It is also a social identity, a source of dignity and a way through which individuals understand their contribution to the world. When people are told that their work can be automated, they may hear something very different: that the experience they spent years accumulating is becoming less valuable. The distinction between replacing a task and displacing a person's sense of purpose is easy to overlook in a spreadsheet, but it can become enormous in everyday human experience.

None of this means that technological progress should stop. Humanity has never remained stationary, and there is little reason to imagine that artificial intelligence will be the final transformation of work. Some occupations will disappear, others will change, and new possibilities will emerge. The more difficult responsibility lies in deciding how these transitions are managed. Organizations that introduce AI without considering the psychological consequences may intensify the insecurity they are trying to manage. Education, professional retraining, transparent communication and meaningful participation in technological decisions are important not simply as employment policies, but as ways of maintaining the social conditions necessary for progress.

There is something extraordinary about this moment in our technological history. A biological species shaped by natural selection has created computational systems capable of performing increasingly sophisticated intellectual tasks. We are now examining the consequences of that achievement, not only for machines and markets, but for our own psychology. Perhaps the greatest challenge of artificial intelligence will not be determining how much work can be transferred to machines. It will be understanding what happens to human identity when intelligence is no longer a capability that distinguishes us as clearly as it once did. We have spent centuries building machines to extend our abilities. Now we must learn how to live in a world where those machines challenge our assumptions about those abilities. The real measure of our intelligence may ultimately lie not in how successfully we build machines that can perform our work, but in whether we are wise enough to create a future in which technological capability and human dignity can advance together.


Further reading

Probst, T. M., Petitta, L., Roll, L., Lindgren, R. J., Barbaranelli, C., & Ghezzi, V. (2026). “In the Shadow of AI: How Industry Exposure to AI and Occupation Insecurity Fuel Moral Disengagement and Counterproductive Work Behaviors.” Occupational Health Science, 10, Article 27. https://doi.org/10.1007/s41542-026-00272-3.

Share this article
Facebook WhatsApp X LinkedIn Medium
Quick navigationEsc