ECB Finds Only 7% of Euro-Area Firms Use AI Intensively

The European Central Bank published a research blog summarising a firm-level survey of AI adoption across the euro area, which Reuters reports covered more than 5,000 companies. Per Reuters' coverage of the ECB researchers, over 70% of firms report some AI use but only 7% report "intensive" AI use. Reuters quotes the ECB researchers: "The intensive use that drives transformation and generates macroeconomic gains remains rare." PYMNTS' reporting on the ECB blog adds that nearly all intensive users (99%) planned to invest in AI this year and that intensive users earmarked roughly 20% of their total investment to AI-related activities (PYMNTS; ECB blog). The ECB analysis, as reported, finds intensive use concentrated in younger, smaller, and service-oriented, especially high-tech, knowledge-intensive firms (Reuters; ECB blog).

Companies reporting any AI use are not the same as companies that have integrated AI into core processes. Industry-pattern observations show early adopters commonly start with general-purpose licences and point tools; moving to custom integrations typically requires additional data engineering, compute, and specialised staff. Reporting from the ECB blog, as relayed by Reuters and PYMNTS, aligns with that pattern: the blog highlights that intensive users spend heavily on customised solutions rather than only purchasing licences (Reuters; PYMNTS).

Observed patterns in similar diffusion studies indicate that a high headline adoption rate can coexist with low transformative deployment. For practitioners and vendors, this usually means the market is split between opportunistic, licence-based users and a smaller group investing in bespoke systems and infrastructure. The ECB's finding that intensive users allocate about 20% of investment to AI (PYMNTS citing the ECB blog) underscores the capital intensity of deeper integration compared with licence-only adoption.

Industry observers should track three measurable indicators reported by the ECB researchers and press coverage:

• •changes in the share of firms reporting intensive use in subsequent surveys
• •sectoral diffusion beyond high-tech services into manufacturing and large enterprises
• •the composition of AI spending (licensed tools versus customised development and compute). Reporting also notes peer effects: firms invest when competitors do, which suggests diffusion may accelerate once more organisations commit to bespoke deployments (Reuters; ECB blog)

The surveyed results reported here derive from an ECB research blog summarised in Reuters, with additional detail in PYMNTS' coverage of the same ECB post. The Reuters story explicitly states the authors are ECB researchers and that their post "does not necessarily represent the ECB's views" (Reuters).