AI and the Future Job Market

What people think about AI

In recent years, artificial intelligence has fundamentally transformed not just our technological landscape but increasingly our working lives as well. We haven't yet arrived at all-powerful superintelligence, but the effects on society, the labor market, and recruiting are already tangible. In a recently published study, the Gottlieb Duttweiler Institute examined this societal shift in detail.

Through a representative survey across the DACH region — 3,031 consumers and 287 managers with leadership responsibility — GDI asked: What do people think about AI, and what developments do they expect?

The answer is ambivalent. 31 percent of respondents view AI positively overall, while 27 percent lean negative. People feel uncertain when thinking about the future — their feelings are mixed.

When it comes to specific expectations, a more nuanced picture emerges. The biggest changes are anticipated in the workplace: 64 percent expect major upheaval there. Expectations are similarly high for learning and professional development. Somewhat lower, but still significant, are the expected changes in daily life, shopping, and social communication.

Looking more closely at which changes are expected to be positive and which negative, concerns outweigh hopes. Respondents do see some improvements — particularly in the efficiency of public administration. But the worries revolve mainly around money and jobs: 60 percent expect a decline in the number of shops in city centers, and 50 percent expect fewer available jobs.

The Swiss labor market feels the impact

The concern is not unfounded. A recently published study by KOF at ETH Zurich reveals a clear correlation: since the launch of ChatGPT in November 2022, the number of job seekers in AI-affected occupations has risen significantly faster than in less affected occupations — across all age groups.

Programmers and software developers are among the losers. Younger workers are more affected than older ones — though the gap is smaller than in the United States. The question presses: if AI primarily replaces programming and administrative work, which skills are still in demand?

The skill mismatch

A recent report from Wharton University provides answers. The researchers systematically analyzed which skills workers offer on their résumés and which employers demand in job postings. The finding is revealing.

The most in-demand skills are not technical — they revolve around interpersonal abilities. At the same time, many applicants advertise broad, generalist signals like “communication” or “leadership” — skills so widespread that they barely differentiate anymore.

With AI, this becomes even more relevant: routine and standardized work gets automated. What moves to the foreground is whatever is scarce, specific, and value-creating.

What follows from this

Source: Wharton University & Accenture

Soft skills beat technical abilities

The growing importance of social competencies can also be demonstrated empirically. A study published in the Financial Times shows: social skills are now more strongly linked to employment and wages than purely technical abilities. Occupations that demand both high social and high mathematical skills have seen the largest wage increases since 1980.

AI is increasingly automating standardized, quantitative work. The value of human labor is shifting toward collaboration, communication, judgment, and team coordination.

Knowing, Wanting, Acting

In its Future Skills study, GDI developed a model that explains which human capabilities will endure in an AI-shaped future. It rests on three pillars:

Machines are getting better and better at knowing — systems like ChatGPT have essentially read the world's knowledge. And they're also making progress in acting: from agriculture to construction to the latest developments in robotics.

Even though the focus lies on human capabilities, AI competence remains indispensable: a technological understanding, the ability to interpret statistics, and an awareness of how AI impacts one's own work.

AI in HR: The agentic hiring process

A concrete example from Switzerland shows how AI is already affecting daily work in human resources.

AI is making it ever easier to create application dossiers. This lowers the effort barrier to sending out a job application — dramatically. Using ChatGPT to polish a cover letter is just the beginning.

Vibe Application

Derived from the currently popular term “vibe coding” — no longer writing the code yourself, but telling an agent what you want. The same principle already works for the hiring process: tell an AI agent which job you're looking for, and let it do all the work.

Using Claude Code, Anthropic's AI agent, an entire application pipeline can be automated: scanning job profiles automatically, tailoring CVs to each position, leveraging skill matching, sending out mass applications.

What does this mean for HR?

70 percent of companies experimenting with AI are doing so in HR. The most common use cases: writing job postings, automating administrative tasks, and matching candidates by competency.

Agentic Hiring: Agent vs. Agent

The future of recruiting could look like this: AI agents apply autonomously to AI recruiting systems. Agent talks to agent. The efficiency gains are enormous — but the question remains whether a process in which humans only truly meet on the first day of work still leads to the right hires.

The recommendation: prioritize candidate orientation, maintain human interaction, monitor regulatory developments — and invest in upskilling your own recruiters.