Enterprise AI deployments across Europe, the Middle East, and Africa are losing momentum, with new data from IDC pointing to execution gaps rather than fading interest as the primary obstacle.
After an 18-month surge in investment—particularly in large language models and machine learning—many organisations are now slowing or reshaping their strategies. Boards are increasingly demanding clear financial outcomes before approving further expansion, as competing IT priorities and broader economic pressures tighten budgets.
The result: only 9% of organisations in the region have achieved measurable business impact from most of their AI initiatives over the past two years. The remaining 91% are effectively stalled, with projects lingering in pilot phases rather than scaling into core operations.
Rethinking ROI in the AI era
Traditional procurement models are proving ill-suited to evaluate AI investments. Many organisations still measure returns by comparing software costs against potential headcount reductions—a framework that fails to capture AI’s indirect value.
In practice, the benefits of AI often emerge through avoided risks, productivity gains, or new revenue streams. A predictive maintenance system, for example, may not reduce staffing levels but can prevent costly equipment failures—savings that rarely appear in standard financial models.
Without a structured approach to quantifying these gains, promising pilots frequently lose funding before reaching production. CIOs are being pushed to redefine return-on-investment metrics, linking AI performance directly to broader financial outcomes.
Infrastructure gaps slow scaling
Moving AI projects from experimentation to enterprise deployment is proving capital-intensive. While early-stage pilots can run on cloud-based testing environments, scaling requires sustained investment in infrastructure, data pipelines, and ongoing model maintenance.
Integration challenges are particularly acute. Many organisations are attempting to connect modern AI architectures—such as vector databases and retrieval-augmented generation systems—with legacy platforms like on-premise ERP and database environments. Poor data quality further compounds the issue, leading to unreliable outputs and increased error rates.
The cost of restructuring data environments and supporting continuous model inference is rising sharply, forcing CIOs to justify growing cloud expenditures to finance teams.
Regulation as a catalyst—not a constraint
Strict data protection and cybersecurity regulations across Europe are also shaping deployment strategies. Requirements around model transparency, auditability, and protection against threats such as prompt injection attacks are increasing operational complexity.
However, IDC notes that leading organisations are treating compliance as a design advantage rather than a burden. By embedding governance frameworks early, these companies are accelerating deployment timelines while improving resilience, ESG performance, and customer trust.
The human barrier to adoption
Technical readiness alone is not enough. Many AI initiatives face resistance from employees when tools fail to align with existing workflows.
IDC highlights that adoption improves when systems are designed around user needs rather than imposed as standalone solutions. Organisations investing in reskilling and change management are seeing stronger uptake, particularly when AI tools clearly reduce day-to-day friction.
For example, automation in contract review can shift legal teams toward higher-value negotiation work instead of routine compliance checks—making the technology immediately relevant to end users.
CIOs under pressure to deliver growth
The role of the CIO is also evolving. According to IDC, 42% of C-suite leaders in EMEA now expect technology chiefs to lead digital and AI transformation with a focus on generating new revenue streams.
This marks a shift away from traditional IT management toward a more commercially driven mandate. CIOs are increasingly required to tie AI initiatives directly to business performance and ensure alignment across departments.
Execution separates leaders from laggards
As AI adoption enters a more mature phase, success is increasingly defined by execution. Organisations that are progressing beyond pilots share common traits: they align projects with commercial objectives, invest early in governance, and design systems around human workflows.
With financial scrutiny intensifying, the ability to measure returns and build scalable frameworks will determine which companies convert AI investment into tangible value.
