Help, AI is coming: Do I still have a professional future?

A brief journey through fear, change, and reality - in search of an answer to whether we and future generations still have a professional future, or whether we’ll soon find ourselves out of work due to AI and automation.

My For-You Page seems determined to scare me, as an AI guru shouts, “Ninety-five percent of the jobs that exist today will disappear within ten years!” while I’m still trying not to spill my coffee on the keyboard in shock. 

I catch myself thinking: What if AI eventually becomes better at operating AI than I am? Then I wouldn’t just be an observer; I’d be part of the disruption menu myself. 

To calm down, I can’t help but think of my parents’ generation, who must have experienced something similar when computers and the internet first entered everyday life. 

So I began to dig into the past, the present, and a bit into the future. Not out of nostalgia, but out of curiosity. I wanted to know whether the fear of losing our jobs is justified this time, or if we’re simply stumbling into an old story once again. 

Looking Back: We’ve Been Here Before

The fear of becoming redundant is older than any machine. In the 1920s, coachmen feared the automobile; later, office workers feared the computer. Every technological wave brought both displacement and “birth.” An article from the World Economic Forum puts it aptly:

“In a way, the coachmen were right: cars did in fact replace horse-drawn carriages. But their children and grandchildren found new, and often better-paid, jobs in all the activities that only became necessary or possible because of the car: automobile production, vehicle repair, travel services, home delivery, mass tourism, road construction, the gasoline business, and much more.” 

The American economist Joseph Schumpeter once called this creative destruction, a process that replaces the old to make way for the new. The World Economic Forum estimates that 65 percent of today’s children will work in professions that don’t yet exist. No coachman could have imagined what a UX designer is, and no accountant could have guessed what prompt engineering means. Yet these are precisely the kinds of jobs filling today’s job boards. 

Forecasts: When Models Try to Understand the Future 

Of course, no one can seriously predict exactly how AI will change the labor market. But there are attempts to narrow it down. 

Eloundou, Manning, Mishkin, and Rock (2023) examined the potential impact of large language models (LLMs) on the U.S. labor market. Their results show that for about 80 percent of all employees, roughly 10 percent of their tasks could be fully automated, while for 19 percent of employees, up to half of their tasks could be automated. According to the researchers, the most affected are no longer low-skilled jobs, as in the past, but increasingly academic professions. 

Other studies, such as Felten et al. (2023) and Briggs et al. (Goldman Sachs, 2023), confirm this shift but point to crucial variables: the speed at which workers adapt and the actual capabilities of generative AI over time will determine whether, and above all how quickly, automation and AI replace or complement jobs. 

Note: as already mentioned, the predictive power of these models is limited. The numbers are based on data from the Bureau of Labor Statistics and European working-time statistics, but the models combine many estimates and factors (digitalization levels, technical automation potential, etc.). 

Germany’s Federal Ministry for Economic Affairs found that between 2016 and 2018, the early use of AI alone created around 48,000 new jobs. The key point is that AI does not replace work; it replaces tasks, thereby creating new forms of employment. 

Augmentation Instead of Automation 

Humans remain part of the equation. Almost all models assume that most occupations will not disappear due to AI but will change. The keyword here is augmentation, the enhancement of human abilities through technology. 

If ChatGPT does the writing for me, that doesn’t free me from responsibility. On the contrary, I have to check, compare, and interpret. The internet gave us access to an enormous amount of information, and AI now helps us process it even faster. The meaningful, contextual use of that information remains open, and thus emphasizes the importance of humanity, the “human in the loop” within this value chain. 

The Present: Between Black Box and Reality 

Generative AI is still far from replicating the complex transfer processes of human thinking. Drawing conclusions from experience, setting expectations for answers, and shaping those answers into contextually relevant conclusions remains too great and multifaceted a task for language models. 

This is not only due to computing power but also to the opacity of their internal logic. The famous “black box” remains closed: we don’t know why a model responds the way it does. Research fields such as Explainable AI are trying to change this, so far with limited success. 

Conclusion: Between Repetition and Progress 

There are good reasons to believe that the current technological transformation resembles those of past eras. Once again, we are standing at a threshold where work does not disappear but shifts. The history of industrialization, electrification, and digitalization repeats itself, though this time with a tool that automates not only power but also language. 

Reliable forecasts of the long-term effects remain an illusion. Too many variables lie beyond calculation: the speed of adaptation, social acceptance, and the political framework. Yet the trend clearly points to a redistribution of activities, away from purely repetitive tasks toward higher-value, analytical, and creative work. On a large scale, this will likely not mean a replacement of humans but an extension, augmentation instead of automation. 

This also raises expectations for workers themselves. Those who work in the future will need to combine, evaluate, and transfer more. This ability, the linking of knowledge, context, and intuition, remains for now a deeply human asset. It keeps us at the center of the value chain. 

Because AI learns from everything we have already created, its learning models are ultimately a mirror of our language, our culture, and our decisions. In this sense, the data it learns from mark both the natural boundary of its development and the proof that progress always rests on human experience.