European firms push on with AI pilots even as payoff doubts grow

IDC and Lenovo say enterprises are pressing ahead with pilot deployments despite mixed evidence on returns

Despite a growing number of reports that AI is not benefiting many businesses, Lenovo and IDC say that firms in EMEA are pushing ahead with pilot deployments and still expect it to drive growth and transform how they operate.

Lenovo's newly-published CIO Playbook 2026 claims that enterprises across Europe and the Middle East are accelerating their adoption of AI and view it as a key business enabler as they seek speedier growth and better business outcomes.

Based on research carried out by IDC, the report says that some organizations are projecting average returns of $2.78 for every dollar invested in AI projects, but there are considerable challenges to realizing this goal. These include employee skills, having the right infrastructure, robust data security, sovereignty, and governance measures.

The report makes it clear that AI is not just one technology, but an umbrella term that covers things such as predictive and interpretive AI, as well as the generative and agentic types that have been the focus of much hype and attention lately.

IDC's AI Research Director for Europe, Ewa Zborowska, says that compared with 2024, more businesses are now moving beyond the early stages of investigating and playing around with AI, and toward planning and running pilot deployments.

"Almost 60 percent of companies today say they are piloting, they are running proof of concepts, so really seeking value in AI adoption, or already adopting AI systematically, meaning it's part of their business day-to-day," she commented.

Businesses are mostly applying AI in IT-related areas, according to Zborowska, such as cybersecurity or software development, where processes are in place and a lot of data is already available to make use of. "But we are also increasingly seeing customers saying AI is something that is helping us with R&D generally, in operations, customer service, and marketing," she added.

IDC's research covered 800 IT and business decision-makers across the EMEA region.

Almost all of those surveyed (94 percent) expect to see a positive return on the investment they are plowing into these AI initiatives.

However, the reality has proven very different in some recent studies. Research from professional services firm Deloitte showed that adopting AI tools had not helped the bottom line at all for most companies. A PwC survey of 4,454 business leaders found that more than half of CEOs indicated they had seen neither increased revenue nor decreased costs from their AI deployments.

Meanwhile, customers are increasingly looking to non-financial benefits from AI, according to Zborowska.

"So KPIs (key performance indicators) that are not expressed in 'we invested X dollars and we got Y dollars out of this,' they are looking at, are our employees more engaged, are they happy with the tools that they are getting? Has their work improved in any way? How do AI implementations impact our customers' experience, are we making decisions internally in a more efficient way?" she commented, adding that she expected to see non-financial benefits become more prominent in studies over the next few years.

But challenges still remain, including training and upskilling of both IT teams and business users, the report says, and that AI is not going to run well in production without efficient AI infrastructure.

The latter is where Lenovo's interest in the report comes in, of course, as the firm is a global supplier of servers and related infrastructure. IDC placed it joint third in global server market revenue for the third quarter of 2025.

In this study, IDC found that hybrid deployment is the preferred model, with 82 percent of organizations looking to a mix of on-premises or edge infrastructure, with public cloud appearing to be the minority choice. This is a turnaround from model training, where the hyperscale public cloud operators have been among the biggest beneficiaries of AI investment.

More than half of companies reported that they have yet to develop AI governance, risk and compliance policies, or are in the process of doing so. Only 27 percent indicated they have comprehensive guidelines and practices that rigorously address security, data protection, privacy, and AI sovereignty.

For 2026, IDC says it expects to see enterprise spending on AI continue to rise, and that it will become a core part of business performance. The report also says that agentic AI will emerge as the next step in workflow automation, though this needs to be applied with caution. ®