Share of firms using AI in EU strategic sectors rose to 18% in 2025
An analysis by the Joint Research Centre finds uptake grew from 7% in 2021, led by media, energy and electronic communications, while mobility, tourism and agri-food trail.
AI illustrationThe share of firms using artificial intelligence in the EU's 10 strategic sectors rose from 7% in 2021 to 18% in 2025, according to an analysis by the European Commission's Joint Research Centre. Every one of the sectors advanced, but growth was uneven.
The centre said the rise was driven mainly by accessible, general-purpose AI that analyses or generates text, code, images, sound and video. Most of the added use came from systems producing written or audio output for marketing and sales, which have become more accessible and cheaper.
Across the wider EU economy, not only the strategic sectors, adoption was slightly higher, climbing from 8% in 2021 to 20% in 2025. The analysis used Eurostat ICT enterprise survey data gathered between 2021 and 2025.
Cultural and creative sectors and media led with 38% of firms using AI, followed by energy at 34% and electronic communications at 30%. In those sectors firms apply AI to a range of functions, from ICT security to R&D and production. Mobility, tourism and agri-food showed the lowest uptake, with use still limited mostly to marketing, sales and other simple tasks.
Capital-intensive sectors grew faster than those with a more even balance of capital and labour. Uptake among energy firms went from 18% to 34% between 2021 and 2025, and among electronics firms from 12% to 30%.
The centre found that firms making larger structural changes to their organisation to integrate AI are more likely to use it in production. It said that in many cases AI still needs to be built into core processes such as production, where the largest productivity gains are possible. Using off-the-shelf tools does not necessarily lead to deeper integration, the analysis indicates: manufacturing, with low overall uptake, ranks above the climate sector on ready-made solutions.
A lack of relevant expertise was the obstacle most widely cited across the strategic sectors. Scarce specialised skills and regulatory uncertainty, including data protection and privacy concerns, stood out as the main bottlenecks. The centre said policy measures should be tailored to each sector, and that the results will help policymakers set targets and design sector-specific measures under the Apply AI Strategy, launched a year ago.