AI not yet replacing jobs in Europe, say ECB analysts

While artificial intelligence is increasingly used by European companies, there is no clear evidence yet that the technology is replacing workers, with many firms actually more likely to hire additional staff, according to analysis by the European Central Bank (ECB).

The analysis was written by Laura Lebastard, economist in economics, supply side, labour and surveillance, and David Sondermann, deputy head of division in economics, supply side, labour and surveillance, who examined how AI use and investment influence hiring and firing decisions at European companies.

“Artificial intelligence is everywhere, and the workplace is no exception,” the authors wrote, as they explored whether the technology will empower workers or ultimately replace them.

They explained that AI has the potential to significantly influence firms’ production processes and could reshape the labour market in the future.

On one hand, the technology could replace workers and lead to lower employment, while on the other it could boost corporate profits and create entirely new types of jobs that complement the technology.

The debate has intensified in recent months after reports in the United States pointed to thousands of job cuts at companies such as Amazon and Target, where artificial intelligence was cited as a contributing factor.

Against this backdrop, the authors asked whether companies have already begun replacing human workers with AI systems.

To examine the issue, they analysed hiring patterns among European companies, comparing firms that use and invest in AI with those that do not.

Their findings draw on the European Central Bank survey on the access to finance of enterprises (SAFE) conducted during the second and fourth quarters of 2025, which included around 5,000 firms.

The survey revealed that AI use is widespread among European businesses, although actual investment in the technology remains far less common.

Two-thirds of the companies surveyed reported that their employees already use AI tools in their work.

However, only around a quarter of firms said they invest directly in AI technology.

The authors said this difference highlights an important development in the digital economy.

“Firms do not necessarily need to invest in AI in order to use AI technology,” the blog stated.

The availability of widely accessible online tools means the barrier to entry is relatively low, allowing even small companies to adopt AI without major capital investment.

Large companies appear to be leading adoption.

The survey found that almost 90 per cent of businesses with 250 or more employees use AI, compared with 60 per cent of firms employing fewer than ten workers.

The analysis then compared employment trends between companies that use AI and those that do not, taking into account several factors known to influence job creation.

These included firm size, company age, changes in investment, turnover, profitability, economic outlook, expected investment trends, sector and country.

Overall, the results showed no significant difference in job creation or job losses between firms that use AI and those that do not.

However, the picture changed when researchers focused on companies that use AI intensively.

The study found that firms making significant use of AI were about 4 per cent more likely to hire additional employees.

In other words, companies heavily using AI tended on average to hire rather than dismiss workers.

A similar pattern was observed among firms investing in the technology.

Companies investing in AI were nearly 2 per cent more likely to recruit additional staff than those that do not invest in it.

The authors suggested that investment in AI often requires additional employees to operate, maintain and develop the technology.

This effect appeared particularly strong among smaller firms.

According to the analysis, AI investment had little impact on employment levels in large firms, while the positive hiring effect was largely driven by smaller businesses.

Some companies may also be investing in AI as part of strategies to expand production and scale up operations.

The authors said this interpretation is supported by the reasons firms gave for using AI.

Employment growth was mainly driven by companies applying the technology to research and development and innovation, which are key drivers of business expansion.

Although the survey did not specify the exact types of employees being hired, the authors said many of them are likely to be highly skilled workers capable of using or developing AI technologies.

A different pattern was observed among companies using AI primarily to reduce labour costs.

In those cases, firms tended to hire fewer workers and report higher levels of layoffs.

However, this group represented a relatively small share of AI users.

Only 15 per cent of firms that use AI said reducing labour costs was one of their motivations, which the authors said was not enough to offset the broader positive employment effects seen so far.

The blog also examined companies’ expectations for future hiring.

When firms currently using or investing in AI were asked about their plans for the following year, the overall results showed no major differences in hiring intentions compared with other companies.

However, the outlook changed when the analysis focused specifically on future AI investment plans.

Companies planning to invest in AI were more likely to expect employment growth in the coming year, even after accounting for their broader investment plans.

This trend appeared regardless of how large the planned investment in AI might be.

The findings suggest that companies are unlikely to pause hiring simply because they are investing in AI technologies, at least over the next year.

However, the authors warned that the picture could change over a longer time horizon.

They referred to a survey by the ifo Institute, which found that many German companies expect AI to result in some job reductions over the next five years.

For now, the ECB analysis suggests that AI investment and intensive use are not replacing jobs in Europe.

Instead, some companies appear to be recruiting additional workers in order to develop and implement AI technologies while continuing their existing production processes.

Looking ahead over the next year, firms planning to invest in AI still expect to hire more employees than those without such plans.

“While some firms may use AI to replace workers, the average firm is more likely to take on additional staff to enable it to use and invest in AI,” the authors concluded.

The blog also addressed why some studies paint a more negative picture of the technology’s impact on employment.

Research findings differ widely because studies often examine different time horizons, geographical regions and aspects of the labour market.

Comparisons between Europe and the United States are particularly difficult, the authors said, because levels of AI investment, the scale of adoption and the timing of technological deployment vary significantly between the two regions.

Overall, the survey results suggest that AI’s impact on employment in Europe remains positive for now, largely because the technology has not yet fundamentally transformed production processes.

However, the authors warned that the long-term consequences of AI for employment remain uncertain, particularly as the technology continues to evolve and become more deeply integrated into the economy.