Banque de France finds weaker youth hiring in occupations more exposed to generative AI
For workers aged 21–29, permanent-contract hiring in the most AI-exposed occupations was about 28% below its November 2022 level in March 2026, the Banque de France reported.
Event / data period: 2026-03
Published: · Updated:
A Banque de France analysis published on 8 October compares permanent-contract hiring of people aged 21–29 in France by occupations’ exposure to generative AI. Using administrative labour-movement data, it found that hiring in the highest-exposure quintile was about 28% below its November 2022 level in March 2026; the decline in the lowest-exposure quintile was about 8%.
BFMTV covered the findings on 10 October. The Banque de France index measures the technical potential for AI to perform or assist with tasks in an occupation, not firms’ actual use of AI. The central bank said the results do not isolate an effect caused by AI alone; monetary tightening, the post-Covid correction in the technology sector and remote work also affected the period.
Data
Banque de France findings on permanent hiring of young workers
Permanent-contract hiring for people aged 21–29 by occupational generative-AI exposure quintile.
| En yüksek maruziyet beşte birlik dilimi: Mart 2026 CDI işe alımları, Kasım 2022 seviyesinin yaklaşık %28 altında. | Highest-exposure quintile: March 2026 permanent hiring was about 28% below its November 2022 level. | خُمس المهن الأعلى تعرضاً: كان التوظيف الدائم في مارس 2026 أقل بنحو 28% من مستواه في نوفمبر 2022. | Quintile le plus exposé : les embauches en CDI de mars 2026 étaient environ 28 % sous leur niveau de novembre 2022. | Quintil más expuesto: la contratación indefinida de marzo de 2026 estaba cerca de un 28 % por debajo del nivel de noviembre de 2022. | Höchstes Expositionsquintil: Die Einstellungen im März 2026 lagen rund 28 % unter dem Stand vom November 2022. | Quintil mais exposto: as contratações sem termo em março de 2026 estavam cerca de 28% abaixo do nível de novembro de 2022. |
| En düşük maruziyet beşte birlik dilimi: yaklaşık %8 aşağıda. Endeks teknik görev maruziyetini ölçer; gerçekleşen yapay zekâ kullanımını veya nedenselliği göstermez. | Lowest-exposure quintile: about 8% below. The index measures technical task exposure, not observed AI use or causation. | خُمس المهن الأقل تعرضاً: أقل بنحو 8%. يقيس المؤشر التعرض التقني للمهام، لا الاستخدام الفعلي للذكاء الاصطناعي ولا السببية. | Quintile le moins exposé : environ 8 % en dessous. L’indice mesure l’exposition technique des tâches, pas l’usage observé de l’IA ni la causalité. | Quintil menos expuesto: cerca de un 8 % por debajo. El índice mide la exposición técnica de las tareas, no el uso observado de IA ni la causalidad. | Niedrigstes Expositionsquintil: rund 8 % darunter. Der Index misst die technische Aufgabenexposition, nicht beobachtete KI-Nutzung oder Kausalität. | Quintil menos exposto: cerca de 8% abaixo. O índice mede a exposição técnica das tarefas, não a utilização observada de IA nem causalidade. |
Methodology
Source: Banque de France blog post No. 463, published 8 October 2026. In permanent-contract hiring recorded in administrative labour-movement data (Dares MMO) from January 2019 to March 2026, hiring of people aged 21–29 in the highest occupational-exposure quintile was about 28% below its November 2022 level in March 2026; in the lowest quintile it was about 8% below. The exposure index measures tasks that AI could technically perform or assist with, not firms’ actual AI use. The findings do not establish causation; monetary tightening, the post-Covid technology-sector correction and remote work were among concurrent factors. The chart series are logged, seasonally adjusted and Hodrick–Prescott filtered; see the source’s Figure 2 note for scope and method.
Updated:
Subjects and places
Subject proposals
- Artificial intelligenceSuggested
- Subject suggested from the title or summary: generative AIwww.bfmtv.com ↗www.banque-france.fr ↗
Geographic references
Only source-provided geographic references are shown. Opening an external map is optional. Optional story maps use OpenStreetMap. Opening a map sends your IP address and this site’s address to the map provider; location data remains readable as text without loading the map.
- France · Place role: Coverage · Place type: Country or area · Location precision: RegionalEvidence
- Explicit preserved editorial country/region; no city or national centroid inferredwww.bfmtv.com ↗www.banque-france.fr ↗