Automatic translation into Dutch is available for this page. Translate this page
A Parisian streetscape is viewed head on, from the middle of a boulevard, with people, cars and restaurants visible, a colourful graphic overlay and the ETF logo in the middle

PRESS RELEASE - AI may not cause mass job destruction, but could intensify labour market inequalities

New ETF report calls for responsible AI governance and explores impacts of AI on labour markets in developing economies

Turin, 17 April 2026 - Artificial intelligence is already transforming how work is organised, monitored and performed, but its overall impact on jobs, working conditions and inequality remains deeply dependent on context. A comprehensive new report published by the European Training Foundation (ETF) highlights that AI is far more likely to change jobs than eliminate them. It stresses that, without deliberate policy choices, AI could exacerbate labour market polarisation and inequality.

The report was presented today at the Biennale Tecnologia, an event organised by the Polytechnic University of Turin, which explores the relationship between technology, humans and society. Providing a comprehensive review of global research on AI’s effects on employment, working conditions and inclusiveness, the highly detailed analysis places special focus on the implications of AI for developing economies, including countries neighbouring the EU.

“AI is impacting labour markets faster than probably any other technology in decades,” said ETF Director, Pilvi Torsti. “This seminal report shows that changes wrought by AI are more complex than what headlines about lost or threatened jobs might suggest. The effects of AI depend on its deployment. When used to support workers, it can enhance productivity and job quality. When used primarily for control and surveillance, it can negatively reduce autonomy and deepen inequality. Leaving AI development solely to market forces risks undermining decent work. A human-centred approach, grounded in skills development – including AI literacy, ethical and regulatory safeguards, and social dialogue – is essential for AI to support fair and inclusive labour markets.”

Jobs are being transformed, not wiped out

Drawing on a wide range of academic studies, international policy analyses and empirical workplace evidence, the report finds little indication so far of large-scale net job destruction due to AI. Rather, AI is found to be reallocating tasks, automating some functions while complementing others. This task-level transformation has mixed consequences depending on occupation, skill level and work organisation. Even in high-income countries, only around 5% of jobs face a high risk of full automation.

Higher-educated, higher-income, predominantly male workers are more likely to benefit from AI-driven productivity gains, while women, older workers and low-skilled employees with routine and low-autonomy jobs face greater exclusion risks, negative bias and deteriorating working conditions. Entry-level white-collar roles are also particularly vulnerable, raising concerns about disrupted career pathways for young graduates.

Work quality under pressure as algorithmic management spreads

Beyond employment figures, the report stresses that AI has a more immediate impact on job quality. AI-driven systems increasingly influence work pace, autonomy, skill use and performance monitoring, especially through the spread of algorithmic management practices from digital platforms to traditional workplaces. While AI can reduce exposure to physically demanding or repetitive tasks, it often coincides with greater intensity, tighter deadlines and increased surveillance. Strong institutions and worker involvement can mitigate negative effects and steer AI adoption towards job-enhancing outcomes.

Some key statistics:

  • Between 40% and 60% of jobs are expected to undergo significant task changes due to AI.
     
  • Around 79% of European firms use at least one algorithmic management tool.
     
  • About 60% of jobs in high‑income countries are exposed to AI, versus 26% in low‑income countries.
     
  • Even in high-income countries, only around 5% of jobs face a high risk of full automation.

About the European Training Foundation

The European Training Foundation (ETF) is the European Union agency that supports EU neighbouring countries in reforming education, training and labour market systems. The ETF works with nearly 30 countries – from Ukraine to the Western Balkans, from the Southern Mediterranean to Central Asia and the African Union – to strengthen the security and prosperity of Europe and its neighbours.

The ETF Governing Board, chaired by Mario Nava, European Commission Director-General for Employment, Social Affairs and Inclusion, brings together representatives from all 27 EU Member States. The ETF is based in Italy.

Media relations