Artificial intelligence: the unseen tsunami heading for the world of work

AI is no passing fad – it’s an evolutionary step comparable to the advent of electricity or the Internet. The foundations of AI were laid by pioneers like Claude Shannon, the US mathematician who, in 1951, built a mechanical, maze-solving mouse called “Theseus”. In the course of his research, he created the basis for modern neural networks – artificial, typically computerized systems modeled on the human brain, that are now used for everything from detecting fraud to optimizing logistics and making medical diagnoses [1].

Large Language Models (LLMs) like ChatGPT take things a step further by generating text, translating languages and writing code. LLMs are based on the transformer architecture that burst onto the scene in 2017. There were several milestones on the way to the development of today’s LLM architecture: the famous test proposed by Alan Turing in 1950, the first chatbot, ELIZA, created in 1966, and of course Deep Blue’s chess victory in 1997. Now, in December 2025, ChatGPT is revolutionizing our everyday lives while Elon Musk enthusiastically predicts a future where billions of robots produce goods at almost no cost. But for all this recent euphoria, there are also those who warn about the consequences of the coming technological disruption. One such voice is Bill Gates, who predicts that “humans won’t be needed for most things”. And even Shannon warned that “I visualize a time when we will be to robots what dogs are to humans, and I’m rooting for the machines”. Thus, we humans cannot help but ask what the future holds for us. Will we be relegated to the role of spectators in the history of the future? Will we be like horses – once an indispensable working animal but now mostly just there for recreation? There are several studies confirming that this is what many people fear [2].

The personal impacts of disruption: job losses and new horizons

Things are coming to a head in Germany. Even in 2023, an Oxford University study was already predicting job losses in manufacturing and customer service from 2025. And this is indeed confirmed by the latest figures for 2025, as published in PwC’s AI Jobs Barometer, for example. The relentless advance of AI is transforming jobs in German enterprises and raising salaries while also requiring a much more highly-skilled workforce [3]. But in Germany, the boom is stagnating. According to the ifo Institute, businesses expect AI to result in job losses over the next five years [4]. The EY European AI Barometer reveals that 70% of employees are concerned about job losses, and 36% fear for their own job. McKinsey predicts that 24% of jobs will be automated by 2030, with 10 million jobs being lost and just 920,000 new jobs being created in fields including AI development and robotics engineering.

Nevertheless, there are also several positive aspects to this transformation. As well as transforming people’s jobs, generative AI is also creating new, hybrid work models. This means that the current outlook isn’t quite as dystopian as some people fear. The “Workday” blog predicts that new roles will emerge in the context of human-machine collaboration, for instance in the care sector, where robots could support up to 90,000 jobs in the future [5]. The Federal Institute for Vocational Education and Training reports that increased use of AI will result in more complex jobs involving less routine, more creative tasks [6]. The debate is also raging on X (formerly known as Twitter). One user warns that autonomous driving in Germany is failing to take off because of over-regulation and insufficient bandwidth. Another argues that the future of AI robots is in the manual trades: “1,000 robots can acquire 140,000 hours of experience in a single week – more than a person can in 74 years.” Alphabet subsidiary Waymo is already providing 150,000 robotaxi rides a week. Meanwhile, German automakers are falling massively behind, hampered by unnecessary regulation and overly high system development costs. And the picture is similar for administrative jobs – AI is breaking down language barriers and taking over routine tasks, once again threatening to replace huge numbers of human roles.

Social tensions brewing: inequality and ethical questions

The dramatic discrepancy between the 10 million jobs forecast to be lost in Germany by 2030 as a result of AI and robotics and the mere 920,000 jobs expected to be created (author’s own forecast) poses major challenges for our country. The threats to society include unemployment, social tensions and growing inequality, with low-skilled and older workers particularly at risk. The education system is under pressure, and cultural upheaval could challenge traditional work identities. Meanwhile, tech regions reap the economic benefits.

These pressures are only exacerbated by Germany’s demographics. While immigration could in principle mitigate the skills shortage, 42% of 19-25 year-olds with a migrant background currently have no professional qualifications [7]. Early childhood education and retraining are the keys to addressing this issue. AI could provide language support, but more human support is also required in kindergartens [8]. The investment needed to make this happen is going through the roof. 90.5 billion euros will have to be plowed into the German education system by 2030, including 30 billion to retrain two million workers so they are ready for the age of AI [9]. Huge subsidies will also be needed to support chip production (20 billion euros by 2027 [10]). Yet Bertelsmann still warns that instead of booming, the AI job market in Germany is actually stagnating [see 3].

There are also various ethical questions that need to be addressed. Who is liable for AI failures in the care sector? How can we give our lives meaning if we no longer have to work? Some users of X are calling for laws to make human labor mandatory in order to avoid over-reliance on AI and ensure that there is a plan B. AI can support social jobs in the care sector, for example, but should not replace them. Moreover, AI could further drive down wages in low-paying jobs such as cleaning by eliminating entry-level jobs.

Outlook: beating the fear and shaping the future

Survey findings indicate that while many people foresee positive impacts like more efficient working, they also expect threats [11]. The vision is a symbiotic world where AI performs routine tasks, freeing up humans to innovate. We could see robots powered by artificial general intelligence even in 2025. This will be a focus of Germany’s planned AI Innovation HUB (www.KIIHUB.de), which aims to connect AI experts from research, startups and SMEs with the general public. The idea is to discuss key issues relating to the imminent AI revolution with a broad cross-section of the public in order to tackle the growing inequality that will inevitably arise from the ever-faster succession of technological disruptions. The warning signs are clear – WirtschaftsWoche magazine reports that the number of Internet searches for “unemployment” has doubled since 2021 [12].

So what is the solution? Lifelong learning, cultural sensitivity and support from policymakers are all essential. In particular, it is incumbent on government to create conditions that support innovation and to promote startups by removing barriers and bringing legislation up to date with the modern world. Aalen University’s AI research project “KI-Werkstatt Mittelstand” [13] is currently testing collaboration between humans, artificial intelligence and autonomous robotic systems. The findings point to the fact that we are at a critical juncture. AI could free us, or it could divide us. It’s not a question of whether AI is coming, but of how we shape its future development. We must ride the tsunami, rather than allowing it to sweep us away.

[1] Melanie, M. Baum (BIBB) et al.: “Digitalisierung und Wandel der Beschäftigung (DiWaBe 2.0): Eine Datengrundlage für die Erforschung von Künstlicher Intelligenz und anderer Technologien in der Arbeitswelt”, Bundesanstalt für Arbeitsschutz und Arbeitsmedizin, Dortmund, 2025
[2] Ernst & Young Global Limited (EY): “Mehr als ein Drittel der Beschäftigten fürchtet wegen Künstlicher Intelligenz um den eigenen Job”, 2025
[3] Büchel, J. F. Engler, A. Mertens: “KI-Jobs in Deutschland: Stagnation statt Boom”, Bertelsmann Stiftung Gütersloh, 2025, DOI 10.11586/2025025
[4] Wohlrabe: “Mehr KI, weniger Jobs? Was Unternehmen in Deutschland erwarten”, IFO Institut, 2025, ISSN 2700-8371
[5] Valero: “Neue Jobs durch KI: Wie sich die Arbeitswelt der Zukunft verändert”, Workday (blog), 2025
[6] Michael Bloss, Green MEP, November 19, 2025: “Deutschland wird autonomes Fahren wohl verpassen. Zu viel Regulierung, zu wenig Bandbreite, zu wenig Mut.”, post on X, 2.3 million impressions JulianRmm, November 21, 2025: “Ein Team von 1.000 KI-Robotern sammelt in einer Woche ~140.000 Stunden Erfahrung. Das entspricht mehr Erfahrung als ein einzelner Mensch in 74 Jahren Lebensarbeitszeit bekommt.”, post on X, 5 million impressions
[7] Baas: “Bildungsbeteiligung nach Migrationshintergrund”, Statistisches Bundesamt, 2021
[8] Böhmer et al.: “Gelebte Vielfalt: Familien mit Migrationshintergrund in Deutschland”, Bundesministerium für Familie, Senioren, Frauen und Jugend, 2020, article: 2BR193
[9] Estimate based on McKinsey and Federal Ministry of Education and Research data on workers’ training needs.
[10] In 2023, the Federal Government announced 20 billion euros of funding from the Climate and Transformation Fund to strengthen semiconductor production.
[11] de: “Rechnen Sie damit, dass sich künstliche Intelligenz in den kommenden fünf Jahren auf Ihren Arbeitsplatz auswirkt?”, 2025
[12] Weyand: “Die Angst vor der Arbeitslosigkeit kehrt zurück”, WirtschaftsWoche, 2025
[13] www.hs-aalen.de/de/facilities/243