---
title: "Young French Workers Face About 6,200 Fewer Permanent…"
canonical: "https://www.metaintro.com/blog/france-young-workers-6200-fewer-permanent-hires-month-ai-exposed-jobs-banque-de-france"
language: "en"
author: "drashtigarach"
published: "2026-10-10T08:19:13.000Z"
modified: "2026-10-10T08:19:13.000Z"
---

[Back to Blog](/blog)
[News](/blog/tag/news)[AI](/blog/tag/ai)[Europe](/blog/tag/europe)[Labor Market](/blog/tag/labor-market)[Jobs](/blog/tag/jobs)
# Young French Workers Face About 6,200 Fewer Permanent Hires a Month in the Most AI-Exposed Jobs

About 6,200 fewer permanent hires a month for French workers aged 21 to 29 in the most AI-exposed jobs, Banque de France and HEC economists find. Read more.

[![Drashti Garach](https://cdn.metaintro.com/rs:fill:40:40/q:72/plain/images/5719d740-e510-42bc-8017-e040d145f35f_1766029465094.png)Drashti Garach @DrashtiGarach](/blog/author/drashtigarach)

[October 10, 2026](/blog/archive/2026/10)14 min read

![Blank contract folders and a pen on a pale wooden counter in a bright Paris recruitment office](https://cdn.metaintro.com/rs:fill:1200:675/q:78/plain/images/final-10516.jpg)

[https://x.com/intent/tweet?text=Young%20French%20Workers%20Face%20About%206%2C200%20Fewer%20Permanent%20Hires%20a%20Month%20in%20the%20Most%20AI-Exposed%20Jobs&url=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Ffrance-young-workers-6200-fewer-permanent-hires-month-ai-exposed-jobs-banque-de-france](https://x.com/intent/tweet?text=Young%20French%20Workers%20Face%20About%206%2C200%20Fewer%20Permanent%20Hires%20a%20Month%20in%20the%20Most%20AI-Exposed%20Jobs&url=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Ffrance-young-workers-6200-fewer-permanent-hires-month-ai-exposed-jobs-banque-de-france)[http://www.facebook.com/sharer.php?u=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Ffrance-young-workers-6200-fewer-permanent-hires-month-ai-exposed-jobs-banque-de-france](http://www.facebook.com/sharer.php?u=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Ffrance-young-workers-6200-fewer-permanent-hires-month-ai-exposed-jobs-banque-de-france)[https://www.linkedin.com/sharing/share-offsite/?url=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Ffrance-young-workers-6200-fewer-permanent-hires-month-ai-exposed-jobs-banque-de-france&title=Young%20French%20Workers%20Face%20About%206%2C200%20Fewer%20Permanent%20Hires%20a%20Month%20in%20the%20Most%20AI-Exposed%20Jobs](https://www.linkedin.com/sharing/share-offsite/?url=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Ffrance-young-workers-6200-fewer-permanent-hires-month-ai-exposed-jobs-banque-de-france&title=Young%20French%20Workers%20Face%20About%206%2C200%20Fewer%20Permanent%20Hires%20a%20Month%20in%20the%20Most%20AI-Exposed%20Jobs)[mailto:?subject=Young%20French%20Workers%20Face%20About%206%2C200%20Fewer%20Permanent%20Hires%20a%20Month%20in%20the%20Most%20AI-Exposed%20Jobs&body=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Ffrance-young-workers-6200-fewer-permanent-hires-month-ai-exposed-jobs-banque-de-france](mailto:?subject=Young%20French%20Workers%20Face%20About%206%2C200%20Fewer%20Permanent%20Hires%20a%20Month%20in%20the%20Most%20AI-Exposed%20Jobs&body=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Ffrance-young-workers-6200-fewer-permanent-hires-month-ai-exposed-jobs-banque-de-france)

Permanent (CDI) hiring of French workers aged 21 to 29 in the jobs most exposed to generative AI stood [about 28% below its November 2022 level in March 2026, against 8% in the least exposed jobs](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes), a gap of roughly 20 percentage points. That fall equals [about 6,200 fewer permanent hires a month in the most exposed fifth, of which about 4,600 go beyond the decline in the least exposed jobs](https://heclift.github.io/website/files/papers/LIFT_PolicyPaper_2026-02.pdf). The research is by economists from the Banque de France, HEC Paris's LIFT lab and INSEAD, [carried out in collaboration with the Banque de France](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf), and was [published on October 8, 2026](https://www.epochtimes.fr/ia-les-embauches-en-cdi-des-jeunes-reculent-de-283-dans-les-metiers-les-plus-exposes-3365491.html). For [Metaintro](https://www.metaintro.com) readers starting a career in office, data or software work, the useful part is the detail behind the headline number, covering which jobs count as exposed, where the 75,000 figure comes from, and why the authors say their analysis cannot attribute the drop to AI.

## What Did the Banque de France and HEC Study Actually Find?

The findings appear in three documents. One is [Banque de France blog post number 463](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes), the second is the longer [LIFT working paper 2026-02, dated October 2026](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf), written by Marco Guido Palladino, Antonin Bergeaud, Antoine Bertheau, Simon Bunel and Oscar Degos, and the third is a short [LIFT policy paper 2026-02](https://heclift.github.io/website/files/papers/LIFT_PolicyPaper_2026-02.pdf) that summarises the main results.

The researchers tracked [permanent contract hires and contract endings separately between January 2019 and March 2026](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes), and used November 2022, [the month of the public launch of a popular AI chatbot](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf), as the reference point. Since then, both permanent hires and permanent contract endings have fallen more in the most exposed jobs, [especially among young people](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf).

The key comparison is between the top and bottom fifths of jobs ranked by exposure. For workers aged 21 to 29, CDI hires in the most exposed fifth were [about 28% below their November 2022 level in March 2026, against about 8% in the least exposed fifth](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf). The authors call the overall picture a kind of freeze, with [fewer entries and fewer exits for new entrants in exposed jobs](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf).

The records come from employer payroll declarations and cover [private sector establishments in mainland France, excluding agriculture, temporary agency work and household employers](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf).

## Where Does the 75,000 a Year Figure Come From?

A figure of 75,000 fewer permanent contracts leads [some coverage](https://www.seneweb.com/en/news/International/75-000-cdi-en-moins-comment-lia-freine-lembauche-des-jeunes-en-france_n_999000163.html). It is [the authors' own yearly version of the 6,200 a month figure, for 21 to 29 year olds only](https://heclift.github.io/website/files/papers/LIFT_PolicyPaper_2026-02.pdf), printed in their short policy paper. No 75,000 figure appears in the [blog post](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes), while [about 6,200 fewer CDI hires a month in the most exposed fifth](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf) appear in the working paper.

75,000 is [an indicator of fewer hires, not to be confused with as many jobs cut](https://www.epochtimes.fr/ia-les-embauches-en-cdi-des-jeunes-reculent-de-283-dans-les-metiers-les-plus-exposes-3365491.html) (translated).

The first changes run [more through entries and exits than through a visible fall in the total number of employees](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes) (translated).

## How Was AI Exposure Measured, and Which Jobs Count as Most Exposed?

The exposure index starts from the ROME catalogue built by France Travail, which [describes each occupation by the skills it uses](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes). The [ROME job reference from France Travail](https://www.francetravail.fr/employeur/vos-recrutements/le-rome-et-les-fiches-metiers.html) groups jobs into fiches that cover the definition, access to the job, skills, work contexts and career mobility, and each fiche carries a five-character code.

A language model scored how far AI can perform or assist each skill, taking into account [what limits substitution, such as tacit expertise, physical presence or human interaction](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes). The fiches draw on [16,740 distinct skills, each assessed once](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf). The index measures [potential technical capability, not the actual use of AI by companies](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes) (translated), although its correlation with Google Atlas data on actual AI use [is high](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes).

The most exposed fifth includes [word processing operators, economic, financial and commercial research managers, interpreters and translators, journalists and bank operations technicians](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf). What they share is that they [produce, process or format information](https://www.journaldunet.com/business/action-publique/1556107-les-embauches-en-cdi-des-jeunes-actifs-reculent-dans-les-metiers-exposes-a-l-ia-generative/) (translated). At the other end sit jobs [built on physical presence or direct contact, such as nurses, childminders and skilled workers](https://www.journaldunet.com/business/action-publique/1556107-les-embauches-en-cdi-des-jeunes-actifs-reculent-dans-les-metiers-exposes-a-l-ia-generative/) (translated).

There is a catch in how the groups are built. Each fifth is a set of occupations, and a fifth of occupations [does not necessarily correspond to one fifth of employees](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf).

## Why Does the Gap Shrink With Age?

The gap is [largest for the youngest group](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes) and fades with age. After the 21 to 29 group, it [narrows for 30 to 39 year olds and 40 to 49 year olds, then becomes almost non-existent for employees aged 50 and over](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes) (translated). Those gaps come out at [20.8 points for 21 to 29 year olds, 18.6 for 30 to 39 year olds, 15.3 for 40 to 49 year olds and 3.8 after age 50](https://heclift.github.io/website/files/papers/LIFT_PolicyPaper_2026-02.pdf).

The reasoning offered is about tasks and experience. At a given job, early career workers may be the most vulnerable because they [do more of the tasks generative AI does well and do not yet have the experience that makes an employee harder to replace](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf) (translated).

French labour law [can influence the form this adjustment takes](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf). A firm that expects to need less labour [can limit new hires or leave some departures unreplaced without dismissing current staff](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes) (translated), a margin the authors call [particularly relevant in France, where ending a CDI is more regulated than in several other countries](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes) (translated). So [hiring can react before total employment does](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes) (translated).

> Asriyan et al. (2026) show that automating junior tasks can ... reduce long-term output when the cost savings it brings remain modest, since by closing the first rung of the ladder it interrupts the accumulation of experience on which more skilled jobs rest, [the LIFT working paper](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf) notes (translated).

The French authors add that [their results do not establish that this mechanism is already at work in France](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes) (translated).

## Are Young Workers Also Leaving Their Jobs Less?

Yes, and that is the second half of the freeze. For 21 to 29 year olds, CDI contract endings were [about 23% below their November 2022 level in the most exposed fifth, against 6% in the least exposed](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes). The fall in endings is [smaller than the fall in hires](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf), across the same reference period.

Because hiring fell further than departures, the [balance between hires and endings for 21 to 29 year olds in the most exposed fifth shrank](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf) against the same reference point. The authors describe the freeze as asymmetric, reflecting [both lower labour turnover and weaker workforce renewal](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf).

Firms [do not appear to have replaced permanent contracts with fixed-term ones](https://www.journaldunet.com/business/action-publique/1556107-les-embauches-en-cdi-des-jeunes-actifs-reculent-dans-les-metiers-exposes-a-l-ia-generative/) (translated). Across all contract types, hiring of 21 to 29 year olds in the most exposed jobs fell [22.1% between November 2022 and March 2026, while it rose 1.0% in the least exposed fifth](https://heclift.github.io/website/files/papers/LIFT_PolicyPaper_2026-02.pdf). For a young job seeker, the study describes [fewer entries and fewer exits](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf) in exposed jobs.

## What Happened to Young Developers and IT Engineers?

Developers and IT engineers, defined by the French occupation codes [388A and 478A, which make up a large share of flows in the most exposed fifth](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf), get a closer look.

For 21 to 29 year olds in these jobs, CDI hires in early 2026 were [about 42% below their November 2022 level, against about 11% for the comparison group](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf), a gap close to 31 percentage points. The gap is smaller at older ages but [present in every age group](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf), which shows a slowdown that is not limited to new entrants.

The authors say the developer case [cannot isolate the role of AI from the business cycle in IT activities](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf), marked notably by reorganisations after the pandemic.

## Do Companies That Use AI Hire Fewer Young People?

The researchers matched their data to the Banque de France business survey and followed [about six thousand companies, of which nearly 600 declared intensive AI use in early 2026](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes). The question on AI use was asked [only once, at the start of 2026](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf), with five possible answers ranging from no use to intensive use.

CDI hiring of 21 to 29 year olds has [developed less favourably since late 2022 at intensive users than at other companies](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes) (translated), while the gap is much less clear at older ages. Their total headcount, however, shows [no statistically clear divergence](https://www.epochtimes.fr/ia-les-embauches-en-cdi-des-jeunes-reculent-de-283-dans-les-metiers-les-plus-exposes-3365491.html) from other firms.

Selection matters here. The intensive users were [already recruiting into more exposed occupations before the chatbot launch, even within the same sector](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes), so AI adoption is not spread at random across firms. The adjustment the authors see is mainly in [the mix and pace of hiring rather than a visible shrinking of total staff](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes) (translated).

## Can the Researchers Say AI Caused the Drop?

No, and they are explicit about it. Like most early studies abroad, their analysis [remains descriptive and is not enough to attribute the observed changes to AI](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf) (translated). The firm comparison shows [an association between declared adoption and hiring, without yet separating the effect of AI from the selection of firms that adopt it](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes) (translated).

Several forces [hit the same jobs over the same period](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes):

- **Interest rates.** The European Central Bank raised its deposit rate [from -0.5% to 4% between July 2022 and September 2023](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf), and this tightening, together with the post-Covid correction, [may have weighed on investment and hiring in the same jobs as AI, notably in technology](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf) (translated).
- **The tech correction.** Monetary tightening and the [post-Covid correction in the technology sector weighed on hiring from 2022](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes), including in jobs that are themselves highly exposed to AI.
- **Remote work.** Telework [may also reduce junior hiring and largely concerns the same categories of jobs](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes), and a separate study cited by the authors finds that [firms that adopted telework more then invested faster in generative AI](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf) (translated).

A separate shift sits alongside these. The number of apprentices in France [more than doubled in a few years, passing one million at the end of 2023](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf), the authors note, citing Insee, and this expansion [may influence the trends seen, without on its own explaining differences by exposure](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf) (translated).

The authors argue these are not necessarily competing explanations, since AI, remote work and the hiring cycle [may have affected the same firms and the same tasks together](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes). Next steps include a survey that can [date the year each firm adopted AI](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf).

> While interest rates affect overall employment, existing evidence does not suggest they are a good explanation for the disproportionate decline in entry-level hiring in AI-exposed occupations, Erik Brynjolfsson, Bharat Chandar and Ruyu Chen wrote in February, according to [the Stanford Digital Economy Lab note](https://digitaleconomy.stanford.edu/news/canaries-interest-rates-and-timinga-more-on-recent-drivers-of-employment-changes-for-young-workers).

## How Does France Compare With the United States, Norway and Denmark?

The [results abroad are still mixed](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes). In the United States, a revised paper by Erik Brynjolfsson, Bharat Chandar and Ruyu Chen at Stanford uses ADP payroll data through June 2026 and finds that employment of workers aged 22 to 25 in AI-exposed occupations [now stands 19% below where it would be had it kept pace with less exposed peers](https://digitaleconomy.stanford.edu/publication/canaries-in-the-coal-mine-six-facts-about-the-recent-employment-effects-of-artificial-intelligence/), with no comparable gap for experienced workers. [Like the French study](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes), it finds the change works [primarily through reduced hiring of young workers rather than increased separations](https://digitaleconomy.stanford.edu/publication/canaries-in-the-coal-mine-six-facts-about-the-recent-employment-effects-of-artificial-intelligence/), and it finds [no evidence of widespread, economy-wide job displacement](https://digitaleconomy.stanford.edu/publication/canaries-in-the-coal-mine-six-facts-about-the-recent-employment-effects-of-artificial-intelligence/).

With the broadest set of controls, the Stanford team found suggestive evidence that the decline in AI-exposed jobs [becomes significant only in 2024](https://digitaleconomy.stanford.edu/news/canaries-interest-rates-and-timinga-more-on-recent-drivers-of-employment-changes-for-young-workers), and the earlier declines are likely at least partly due to other factors, as set out in a February 9, 2026 note.

Norway looks different. There, research by Hernæs and Kostøl [finds no relative decline in the most exposed fifth when looking at all young workers' jobs](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes), although employment per head of young developers [falls by about 18%](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf). In Denmark, Humlum and Vestergaard mainly see [a fast reorganisation of tasks, with no detectable average effect on hours worked or earnings](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes) (translated). The French authors read these differences as a reason to look at [occupations, adjustment margins and institutions rather than search for one single average effect of AI](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes) (translated).

## What This Means for Young Job Seekers, and What Should You Do Next?

The French data suggest [a reorganisation of entry into work more than an immediately visible destruction of the total stock](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf) (translated). Tracking total employment alone [could miss part of the adjustment if AI first changes workforce renewal and the training of workers](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes) (translated).

- **Know where the door is narrowest.** The steep drop is in the most exposed fifth of jobs, which includes [interpreters and translators, journalists and bank operations technicians](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf), and it is steeper still for [developers and IT engineers aged 21 to 29, whose CDI hires in early 2026 were about 42% below November 2022](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf). If you are aiming at these jobs, it makes sense to widen your target list to related roles rather than [waiting out a weak first job market](https://www.metaintro.com/blog/weak-first-job-market-graduate-wage-scarring-2026).
- **Look at the skills behind your target job.** The exposure score penalises skills that need [tacit expertise, physical presence or human interaction](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes). Each ROME fiche lists the [skills, work contexts and mobility paths for a job and can be downloaded on MétierScope](https://www.francetravail.fr/employeur/vos-recrutements/le-rome-et-les-fiches-metiers.html), which is a practical way to see which parts of a role lean on judgment and contact.
- **Pitch yourself on what AI does poorly.** Human interaction is [one of the limits on substitution](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes) built into the exposure score. Lead your CV and interviews with examples of judgment and client contact, and show [how you work alongside AI tools](https://www.metaintro.com/blog/early-career-workers-thrive-with-ai).
- **Ask about training.** AI may first change [workforce renewal and the training of workers](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes), and recent theoretical work cited by the authors finds that automating beginners' tasks [can reduce learning by doing](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes), so ask in interviews how juniors learn and who mentors them.
- **Do not read the headline as proof that AI caused it.** The authors say their comparison [does not neutralise monetary tightening, the post-Covid tech correction or remote work](https://heclift.github.io/website/files/papers/LIFT_PolicyPaper_2026-02.pdf).

## Related Articles

- [Why Cutting Entry-Level Jobs Could Backfire on Employers](https://www.metaintro.com/blog/why-cutting-entry-level-jobs-could-backfire-on-employers)
- [Junior Bankers at UBS Must Now Prove They Can Use AI to Get Hired](https://www.metaintro.com/blog/ubs-junior-bankers-prove-ai-skills-hiring)
- [Kia Will Test Graduate Applicants on How Well They Work With AI](https://www.metaintro.com/blog/kia-graduate-hiring-ai-collaboration-assessment)
- [Employers Say New Grads Arrive Missing the Basics and Here Is What to Learn First](https://www.metaintro.com/blog/new-graduates-missing-basics-skills-based-hiring-what-to-learn)
- [AI Is Hiding a Learning Debt That Could Stall Your Career](https://www.metaintro.com/blog/ai-hiding-learning-debt-could-stall-your-career)
- [Why Mentoring Matters More Than Ever in the AI Era](https://www.metaintro.com/blog/why-mentoring-matters-more-ai-era)
- [Australia Says AI Has Not Triggered Mass Layoffs Yet and the Data Backs It Up](https://www.metaintro.com/blog/australia-ai-no-mass-layoffs-2026)
- [Entry-Level Grads Are Grabbing Any Paycheck as the Job Market Turns Brutal](https://www.metaintro.com/blog/entry-level-grads-any-paycheck-brutal-market-2026)
- [IBM Is Betting on Entry-Level Workers as Rivals Cut Them](https://www.metaintro.com/blog/ibm-betting-entry-level-workers-rivals-cut-2026)

## People Also Asked

### Q: Is AI destroying jobs for young people in France?

A: The study does not show job destruction, it shows fewer hires, with first changes running [more through entries and exits than through a visible fall in total employees](https://www.banque-france.fr/fr/publications-et-statistiques/publications/ia-generative-les-premiers-changements-du-marche-du-travail-sur-les-jeunes), and [the authors say their analysis is not enough to attribute the drop to AI](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf).

### Q: Is the 75,000 fewer permanent contracts figure from the study?

A: Yes, the authors' [policy paper converts about 6,200 fewer CDI hires a month into about 75,000 fewer permanent contracts over a year for 21 to 29 year olds](https://heclift.github.io/website/files/papers/LIFT_PolicyPaper_2026-02.pdf). It counts fewer hires, not jobs cut.

### Q: Which jobs saw a steep drop in hiring for young workers?

A: Developers and IT engineers aged 21 to 29 saw CDI hires [about 42% below their November 2022 level in early 2026](https://heclift.github.io/website/files/papers/LIFT_WP_2026-02.pdf), and the most exposed group also includes [interpreters, translators, journalists and bank operations technicians](https://www.journaldunet.com/business/action-publique/1556107-les-embauches-en-cdi-des-jeunes-actifs-reculent-dans-les-metiers-exposes-a-l-ia-generative/).

Future-proof your career. [Metaintro](https://www.metaintro.com) brings together 50M jobs in one place. [Create a free profile](https://www.metaintro.com/signup) and start your search today.

### Share this article

[https://x.com/intent/tweet?text=Young%20French%20Workers%20Face%20About%206%2C200%20Fewer%20Permanent%20Hires%20a%20Month%20in%20the%20Most%20AI-Exposed%20Jobs&url=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Ffrance-young-workers-6200-fewer-permanent-hires-month-ai-exposed-jobs-banque-de-france](https://x.com/intent/tweet?text=Young%20French%20Workers%20Face%20About%206%2C200%20Fewer%20Permanent%20Hires%20a%20Month%20in%20the%20Most%20AI-Exposed%20Jobs&url=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Ffrance-young-workers-6200-fewer-permanent-hires-month-ai-exposed-jobs-banque-de-france)[http://www.facebook.com/sharer.php?u=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Ffrance-young-workers-6200-fewer-permanent-hires-month-ai-exposed-jobs-banque-de-france](http://www.facebook.com/sharer.php?u=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Ffrance-young-workers-6200-fewer-permanent-hires-month-ai-exposed-jobs-banque-de-france)[https://www.linkedin.com/sharing/share-offsite/?url=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Ffrance-young-workers-6200-fewer-permanent-hires-month-ai-exposed-jobs-banque-de-france&title=Young%20French%20Workers%20Face%20About%206%2C200%20Fewer%20Permanent%20Hires%20a%20Month%20in%20the%20Most%20AI-Exposed%20Jobs](https://www.linkedin.com/sharing/share-offsite/?url=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Ffrance-young-workers-6200-fewer-permanent-hires-month-ai-exposed-jobs-banque-de-france&title=Young%20French%20Workers%20Face%20About%206%2C200%20Fewer%20Permanent%20Hires%20a%20Month%20in%20the%20Most%20AI-Exposed%20Jobs)[mailto:?subject=Young%20French%20Workers%20Face%20About%206%2C200%20Fewer%20Permanent%20Hires%20a%20Month%20in%20the%20Most%20AI-Exposed%20Jobs&body=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Ffrance-young-workers-6200-fewer-permanent-hires-month-ai-exposed-jobs-banque-de-france](mailto:?subject=Young%20French%20Workers%20Face%20About%206%2C200%20Fewer%20Permanent%20Hires%20a%20Month%20in%20the%20Most%20AI-Exposed%20Jobs&body=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Ffrance-young-workers-6200-fewer-permanent-hires-month-ai-exposed-jobs-banque-de-france)

For job seekers

## Your next role is already posted.

Say what you want next. We search every day and send real, open roles, with help getting ready for each.

[Get Started Free](/signup)[Search matching jobs](/jobs/search)

- MatchCompare live roles against your current evidence.
- PositionTurn proof projects into role-specific applications.
- ImproveUse market feedback to keep the skill plan current.

[Return to navigation](#main-navigation)