AI could steal Jobs, but the Debate is More Complex than It seems

“Will Artificial Intelligence take my job?” Few questions capture so well the mix of excitement and concern surrounding the current evolution of technology. As AI tools begin to write texts, analyze documents, generate images, program systems, serve customers, and execute entire processes, it is natural to wonder just how far this transformation can go. 

The answer, however, can hardly be reduced to a simple “yes” or “no.” Some jobs will probably disappear. Others will be profoundly transformed. New roles will emerge. And the impact will vary greatly depending on the profession, industry, company, level of experience, and, above all, the type of activity each professional performs. 

Perhaps the more useful question, then, is not simply whether AI will take jobs. The real question is: which types of work are becoming more valuable, and which are becoming increasingly easy to automate? 

The first signs are already emerging 

The concern is not limited to predictions. An update published in August 2026 by the Stanford Digital Economy Lab analyzed payroll data from millions of U.S. workers and found a significant phenomenon. Among professionals aged 22 to 25 in occupations highly exposed to AI, employment levels were approximately 19% lower than they would have been if they had followed the trend seen among young people in less exposed professions. The effect appeared mainly in reduced hiring rather than a wave of layoffs. The researchers themselves, however, caution that the data are descriptive signals and do not yet prove that AI is the sole cause of this difference. 

That caveat is essential. Interest rates, economic slowdown, post-pandemic changes, business strategies, and other factors also affect the labor market. Even so, the pattern deserves attention: the effects appear to be stronger precisely in roles where AI can replace tasks, while occupations in which it serves as a complement to workers show more favorable outcomes. 

A global survey of CEOs cited by Gizmodo points in the same direction. The share of executives who said they intend to reduce junior positions in the coming years reached 43%, while preference for mid-level professionals increased. Interestingly, the same survey shows that companies further along in AI adoption do not always follow this logic: some have begun to see entry-level professionals as more valuable when they are able to use the technology productively. In other words, even among companies making major bets on AI, strategies differ. 

Jobs can disappear without employment disappearing 

It is important to separate two ideas that are often mixed together in this debate. A technology can eliminate certain occupations without necessarily causing a permanent decline in the total number of jobs. This has happened in several previous technological transformations. Machines eliminated manual activities, computers reduced countless administrative tasks, and the Internet made some business models obsolete while creating entirely new markets. AI may repeat this pattern, although potentially at greater speed and scale. 

The International Labour Organization estimates that approximately one in four workers worldwide is in an occupation with some degree of exposure to generative AI. Even so, the study’s conclusion is more measured than many alarming headlines suggest: in most cases, the trend is toward the transformation of roles rather than the complete replacement of workers. 

The World Economic Forum’s Future of Jobs Report 2025 offers another view of this apparent contradiction. Considering different economic and technological transformations, the report estimates the creation of 170 million jobs and the displacement of 92 million by 2030, resulting in a net gain of 78 million positions. When the analysis focuses specifically on AI and information processing, the forecast is approximately 11 million jobs created and 9 million displaced. 

This does not mean that someone who loses a role will automatically find one of the new ones. That is precisely one of the greatest challenges of the transition. The jobs created may require different skills, exist in other industries, or emerge in places different from those where reductions occurred. For people in the labor market, therefore, a positive aggregate balance does not eliminate individual risk. 

Software is on the front line 

Few industries illustrate this shift as clearly as software development. Programmers are among the professionals who began using generative AI intensively at an early stage. Today, systems can suggest code, produce tests, explain complex sections, document applications, find errors, create interfaces, and perform a growing number of tasks that once consumed hours of a developer’s time. This inevitably changes the value of certain skills. 

A junior professional whose contribution is limited to receiving a simple specification and turning it into relatively predictable code now faces competition that did not exist just a few years ago. Part of this work can be done very quickly by AI tools, especially when it involves codified knowledge, known patterns, and repetitive tasks. But that does not mean programmers are heading toward extinction. 

In July 2026, the U.S. Bureau of Labor Statistics was still projecting 15.8% growth in the number of software developers between 2024 and 2034, equivalent to more than 267,000 new positions. The same projection points to strong expansion in related fields such as Data Science, Information Security, and Computer Research. 

The paradox is revealing: AI may reduce the need for certain programming tasks while simultaneously increasing demand for professionals capable of building a world that is increasingly dependent on software and AI. What changes is the bar. 

The challenge for junior professionals 

For decades, many careers were built like a ladder. Professionals started by performing relatively simple tasks, learned from more experienced colleagues, accumulated context, and gradually took on more complex problems. AI may remove precisely some of the first rungs of that ladder. 

Andrew McAfee, an MIT researcher, drew attention to this risk when discussing the automation of entry-level jobs. If companies simply eliminate the activities traditionally used to train new talent, they may discover a few years later that they have stopped developing the more experienced professionals they will need in the future. This is a particularly important concern in software. Where will tomorrow’s senior developer come from if companies stop hiring junior developers today? 

One possible answer is that the very meaning of “junior” will have to change. Instead of spending months performing only repetitive tasks before reaching more interesting problems, new professionals may use AI to accelerate this stage and begin working earlier with integration, analysis, architecture, quality, security, product, and business understanding. This increases the potential of an entry-level professional. At the same time, it raises what is expected of them. 

Knowing a programming language will remain important, but it may no longer be enough. The ability to formulate problems, evaluate AI-generated results, recognize when an apparently correct solution contains flaws, understand requirements, discuss decisions, and learn quickly is likely to become even more important. Professionals do not need to compete with AI at what it does best. They need to develop what makes their contribution harder to replace. 

Automating everything can also be a mistake 

There is another side to this debate as well: companies can also make mistakes when adopting AI. The technical possibility of automating a task does not automatically mean that doing so is the best business decision. AI systems make mistakes, require supervision, depend on data quality, and can introduce risks related to security, privacy, intellectual property, compliance, and quality. 

In addition, eliminating positions simply because certain activities have become automatable may produce immediate cost savings while creating a future problem in knowledge and talent development. The CEO survey cited earlier offers an interesting signal. Although many companies are planning leaner structures, more than half of the executives surveyed said it was still too early to know whether AI investments were delivering the expected return, and only 27% said the results had met or exceeded their expectations. 

The transformation, therefore, is not simply about replacing people with algorithms. The companies that achieve the best results will likely be those able to determine where to automate, where to augment human capabilities, and where to deliberately preserve human participation. 

AI may take jobs, but it can also create work we have not yet imagined 

Denying that AI will eliminate positions would be unrealistic. Some roles are already under pressure, and others will probably be reduced as models and agents become more capable. But claiming that the inevitable outcome will be a world without jobs also ignores the other side of the transformation. 

Developers of AI solutions, data engineers, security specialists, professionals responsible for integrating agents into corporate systems, governance specialists, and countless other activities are growing precisely because of this technology. And many of the roles that will exist a few years from now may not even have established titles yet. 

The history of technology shows that innovation eliminates activities, creates others, and redefines many of those that remain. What is different this time is the speed. Companies and professionals will have less time to react. 

The real risk may be staying the same 

For workers, the message should not be one of panic, but neither should it be one of complacency. The more a role depends exclusively on predictable, repetitive, and easily described tasks, the greater its exposure to automation tends to be. The more it requires context, judgment, applied creativity, relationships, business knowledge, and the ability to use technology to solve problems, the harder it becomes to reduce its value to mere execution. 

For companies, the challenge is similar. Cutting staff may seem like the simplest path when productivity rises, but sustainable transformation requires rethinking processes, roles, talent development, and the very relationship between people and machines. 

After 30 years of following profound changes in the technology industry, Visionnaire understands that the evolution of software has always required adaptation. Artificial Intelligence takes this transformation to another level, creating new possibilities for developing systems, automating processes, and increasing team productivity. 

More than asking whether AI will replace people, companies need to discover how to use it to build better operations, smarter products, and teams capable of delivering what technology alone still cannot solve. The future of work will not be defined only by what AI can do, but by the choices professionals and organizations make with it.