Businesses across Europe, the Middle East and Africa have largely moved beyond debating the importance of artificial intelligence, but many are struggling to deploy it at scale while keeping costs and risks under control, according to market intelligence firm IDC.

The findings, presented by IDC senior research analyst for AI-Fuelled Business Strategies Lapo Fioretti, showed that 99 per cent of EMEA chief executives believed AI could help reinvent how their organisations make money over the next three to five years.

AI strategy has also become one of the two issues, alongside security, that boards increasingly want to discuss beyond financial results.

However, IDC found a significant gap between ambition and capability, with roughly two-thirds of EMEA organisations still at early stages of AI maturity.

The region broadly matched the rest of the world in its appetite for AI but lagged in its ability to implement it effectively.

AI agents were already being used in areas including cybersecurity, IT, human resources and customer service, with 95 per cent of EMEA enterprises reporting that they had agents operating in production.

IDC cautioned that the figure demonstrated the reach of the technology rather than organisational readiness to manage it.

The growing use of AI agents was also creating questions over spending, particularly as businesses struggled to identify costs by individual workflows, lacked dedicated AI financial operations staff and faced opaque pricing models.

At the same time, agent usage across different business functions was expanding faster than governance frameworks.

“Capping spend is not the same as controlling it. That distinction is where the next wave of value, and the next wave of disappointment, will be decided,” Fioretti said.

IDC explained organisations were more likely to advance when they focused AI investment on measurable business value, supported it with funded roadmaps and defined outcomes, and built governance into projects from the outset.

It also identified modular platforms, connected data and continuous learning as common features among organisations making greater progress.

Trust and human judgement remained important despite increasing automation, with reliability ranked as the most important consideration before scaling AI agents, followed closely by accuracy.

Execution speed ranked below both factors, the company added.

“Agentic ROI, in other words, is a discipline – not a calculation,” Fioretti stated.

IDC stated that the combination of high investment ambitions and weaker implementation capabilities could result in smaller deals, longer sales cycles and delayed returns.

The research also pointed to the importance of adapting AI strategies as evidence from deployments accumulates, rather than relying solely on initial forecasts.

For EMEA businesses, the next stage of AI adoption will therefore depend less on deciding whether to invest and more on whether they can establish the financial controls, governance, data infrastructure and human oversight needed to scale the technology.