A striking disconnect is emerging at the heart of enterprise digital strategy. According to new research published by NTT DATA in March 2026, while 99% of organisations acknowledge that artificial intelligence is driving increased cloud investment, only 14% have achieved the highest levels of cloud maturity required to fully capitalise on that investment. For CFOs, CTOs, and boards navigating digital transformation commitments, this gap is not merely operational — it is a strategic liability.

The Infrastructure Deficit Undermining AI Ambitions

The NTT DATA findings reveal a structural misalignment that many executive teams have been reluctant to confront directly: AI ambitions are accelerating faster than the infrastructure required to sustain them. Some 88% of organisations surveyed acknowledged that their current cloud investments risk undermining both AI progress and cloud-native development. The culprit, in most cases, is legacy system debt — inherited architecture that was never designed for the data volumes, latency requirements, or interoperability demands of modern AI workloads.

This challenge is particularly acute for mid-market enterprises across Europe, where cloud maturity levels tend to lag behind large-cap peers. The European regulatory environment — encompassing the EU AI Act, DORA, and NIS2 — adds further complexity, as compliance obligations require robust data governance and auditability frameworks that are difficult to retrofit onto fragmented legacy estates. For General Counsel and compliance officers, the risk is not only operational underperformance but potential regulatory exposure where AI systems cannot demonstrate the traceability and oversight that regulators now expect.

Cloud modernisation has consequently emerged as the top infrastructure priority for the next two years, with organisations urged by analysts to reframe cloud not as a cost centre but as a value-creation platform — a shift that demands new KPIs aligned to business outcomes rather than infrastructure metrics alone.

AI Adoption Is Broad but Depth Remains Elusive

Parallel research from EY reinforces both the scale of AI deployment and the unevenness of its impact. Globally, 78% of organisations now use AI in at least one business function, and 71% report regular use of generative AI. Yet the performance differential between leaders and laggards is stark. EY’s concept of the “superfluid enterprise” — organisations that have systematically scaled AI across functions — shows outcomes that are difficult to ignore: 70% employee adoption rates, 30% reductions in operational cycle times, and returns on AI investment exceeding 150%.

These results do not emerge from isolated pilots. They are the product of deliberate platform strategies, executive sponsorship, and — critically — cloud infrastructure mature enough to support scaled deployment. A 451 Research survey corroborates this, identifying AI integration as simultaneously the top digital workplace priority for 52% of organisations and their single greatest implementation challenge. Integration debt and governance gaps are the primary obstacles, with 58% of respondents indicating a preference for strategic, transformation-oriented IT spending over incremental maintenance investment.

Microsoft’s launch of the Agent Factory framework within Azure AI Foundry signals where enterprise AI is heading: towards orchestrated, multi-agent architectures that automate complex workflows at scale. For organisations still rationalising their cloud estate, this represents a widening competitive gap — not a future concern but a present one.

Implications for Business Leaders and Boards

The convergence of these data points carries clear implications for decision-makers responsible for digital strategy and capital allocation:

  • Cloud modernisation is a prerequisite, not a parallel workstream. Boards approving AI investment programmes should simultaneously mandate a structured assessment of cloud maturity. Without this sequencing, AI budgets risk generating limited returns on inadequate foundations.
  • KPI frameworks must evolve. Measuring cloud performance through cost-per-unit metrics alone obscures the value-creation potential that mature cloud environments unlock. M&A Directors conducting technology due diligence should scrutinise target companies’ cloud maturity as a core value driver, not a secondary consideration.
  • Shadow AI is a governance risk that requires immediate attention. UK research cited alongside the NTT DATA study found that 21% of workers are self-funding AI tools for professional use — a figure that points to unmanaged adoption, data leakage risk, and compliance exposure under frameworks such as the EU AI Act’s transparency and accountability provisions.
  • Regulatory readiness and AI readiness are converging. In the European context, organisations that build compliant, auditable AI infrastructure will be better positioned both commercially and regulatorily as enforcement of the AI Act intensifies through 2026 and beyond.

Key Takeaway

The data is unambiguous: AI value realisation is a cloud maturity problem as much as it is an AI strategy problem. Enterprises that treat digital transformation as a continuous, platform-led discipline — rather than a series of discrete technology projects — are demonstrably outperforming peers across cycle time, adoption, and return on investment. For leadership teams in Europe, the imperative is to close the infrastructure gap with urgency, govern AI adoption systematically, and reframe cloud investment as the strategic foundation upon which competitive differentiation will be built over the next decade.