
An analysis of 7.3 million job advertisements shows entry-level postings down sharply since the years before generative AI, with the steepest declines in roles the tools can most readily take on.
The number of Swiss job advertisements aimed at career starters in 2025 was just under a third lower than the average for the period before generative AI arrived, according to a study published on Wednesday by the recruitment portal Jobs.ch.
The decline was sharpest in the occupations the study labels “AI-exposed,” the white-collar, knowledge-based and office roles whose tasks can be most readily supported or partly automated by the tools.
It is the clearest reading yet of how the Swiss labour market’s lowest rung is thinning.
The figure comes from the company’s AI Report 2026, which drew on 7.3 million advertisements posted on the platforms Jobs.ch, Jobup.ch and JobScout24.ch between 2019 and 2025.
The 💜 of EU tech
The latest rumblings from the EU tech scene, a story from our wise ol' founder Boris, and some questionable AI art. It's free, every week, in your inbox. Sign up now!
The analysts compared postings against a “pre-AI phase” running from 2019 to 2022, the years before ChatGPT and its rivals became workplace fixtures, and supplemented the advertising data with surveys of employees and Swiss firms.
The study covered 18 occupational categories across 19 regions.
Within the AI-exposed sectors, the report found, the share of junior roles fell by 16% relative to the pre-AI baseline, while the share of senior roles rose by 26%.
The affected fields include administration, human resources, banking and finance, marketing, procurement, sales, and information technology and telecommunications.
Taken together, the two movements describe a market that is quietly raising the floor: fewer openings asking for little experience, more asking for a lot.
Jobs.ch is careful about what the numbers prove. The study reads the shift as an indication that companies are either placing greater weight on experience for tasks that were once entry-level, or that those tasks are increasingly being handled with AI tools.
That is a correlation drawn from advertising patterns rather than a measured cause, and the report stops short of claiming AI alone is responsible.
The data captures what employers post, not why they post it. Swiss hiring has also been shaped over the same period by a cooling economy and ordinary caution, neither of which the advertising figures can fully separate out, and the report folds in employee and company surveys partly to read intent the postings cannot show on their own.
The direction of travel, though, matches what other researchers have been describing. A European study found that around 30% of EU workers now use AI at work, concentrated in text-heavy tasks such as writing and translation, and that a large share of firms are reassessing job roles as a result.
The concern in both cases is the same. The routine work that historically gave newcomers a way in, the drafting, the data entry, the first-pass analysis, is exactly the work the tools are quickest to absorb.
That has consequences beyond a single year’s hiring. Entry-level jobs are not only jobs; they are where judgement, adaptability and the habits of a profession are learned. If the rung that used to teach those things is being automated, the people who would have stood on it are expected to arrive already fluent, a tension visible elsewhere too.
American firms have been scaling back internship programmes for similar reasons, and graduates have responded by bringing AI into the interview itself, the very process meant to test them.
The Swiss data does carry one note of nuance the headline can flatten. The contraction is not uniform across all junior work, only across the AI-exposed slice of it, and the report observes that some reshaped entry-level roles are growing rather than vanishing. What is shrinking is the old, undifferentiated bottom rung.
For school-leavers and graduates in administration, finance or marketing, the practical message is blunt enough: the easy first job is getting harder to find, and the one that remains will want more than it used to.
View original source — The Next Web ↗


