The artificial intelligence race among global enterprises has produced a startling finding: despite billions of dollars in investment and relentless executive enthusiasm, only 13% of organizations have reached what could be considered full AI maturity. The gap between ambition and execution is not primarily a technology deficit — it is a human capital crisis that threatens to stall the most significant business transformation in a generation.
That is the central finding from EDB’s 2025 State of AI report, which surveyed enterprise decision-makers across industries and geographies. The report, first covered by Digital Trends, paints a picture of organizations grappling not just with algorithms and infrastructure, but with a fundamental shortage of the skilled professionals needed to deploy, manage, and govern AI systems responsibly.
The Sovereignty Question Is About More Than Data Residency
AI sovereignty — the ability of an organization or nation to maintain control over its AI capabilities, data, and decision-making processes — has become a dominant theme in boardrooms and government ministries alike. But EDB’s research reframes the conversation in a way that many technology vendors have been reluctant to acknowledge. Sovereignty is not simply about where data is stored or which cloud provider hosts your models. It is fundamentally about whether you have the people who understand how to build, operate, and audit AI systems without depending entirely on external vendors or foreign technology providers.
According to the EDB report, the talent dimension of sovereignty is the most underdeveloped pillar across the enterprises surveyed. Organizations may have invested heavily in GPU clusters, data pipelines, and licensing agreements with large language model providers, but they have chronically underinvested in the human expertise required to make those investments productive. The result is a widening gap between the organizations that can truly own their AI futures and those that remain dependent on a small number of hyperscale technology companies.
What the 13% Figure Actually Tells Us
The 13% maturity figure deserves scrutiny. EDB’s methodology assessed enterprises across multiple dimensions, including data readiness, infrastructure, governance frameworks, and workforce capability. To qualify as fully AI-mature, an organization needed to demonstrate proficiency across all of these categories — not just one or two. The low number reflects the reality that while many companies have made progress in isolated areas, very few have built the comprehensive institutional capability required to deploy AI at scale with confidence.
This tracks with findings from other major research efforts. McKinsey’s 2024 Global Survey on AI found that while 72% of organizations reported adopting AI in at least one business function, only a fraction had scaled those deployments enterprise-wide. Gartner has similarly warned that the majority of enterprise AI projects fail to move beyond the pilot stage, often because of organizational and talent-related barriers rather than technical limitations.
The Talent Shortage Is Getting Worse, Not Better
The skills gap in AI is not a new concern, but the nature of the shortage is shifting. In the early years of the machine learning boom, the primary bottleneck was a lack of data scientists and machine learning engineers. Today, the deficit extends far beyond those specialized roles. Enterprises now need AI product managers, data governance specialists, AI ethics officers, MLOps engineers, and domain experts who can translate business problems into AI-solvable formulations. The breadth of talent required has expanded dramatically, and the supply has not kept pace.
EDB’s report emphasizes that this is particularly acute in regions outside the United States, where the concentration of AI talent remains disproportionately high. European and Asia-Pacific enterprises face compounded challenges: they must compete for a smaller local talent pool while simultaneously complying with stricter data sovereignty and AI governance regulations. The European Union’s AI Act, which began phased enforcement in 2024, imposes significant obligations around transparency, risk assessment, and human oversight — all of which require skilled professionals to implement. As Digital Trends noted in its coverage, the regulatory dimension makes the people problem even more urgent for companies operating across multiple jurisdictions.
Enterprises Are Spending on AI, but Not on the Right Things
One of the more provocative implications of the EDB data is that current enterprise AI spending is misallocated. The lion’s share of budgets flows toward technology acquisition — model licenses, cloud computing costs, and data infrastructure — while training, upskilling, and organizational change management receive comparatively modest funding. This imbalance creates a predictable outcome: expensive technology sits underused because the workforce lacks the capability to extract value from it.
The pattern is familiar to anyone who lived through the early waves of enterprise cloud adoption or digital transformation initiatives. Technology procurement consistently outpaces organizational readiness. The difference with AI is that the consequences of this imbalance are more severe. A poorly governed AI system does not just waste money — it can produce biased decisions, expose the organization to regulatory penalties, and erode customer trust in ways that are difficult to reverse.
Why the Database Layer Matters More Than Most Realize
EDB, as a company built around PostgreSQL and enterprise database technology, naturally has a vested interest in emphasizing the data infrastructure layer. But their argument is not without merit. The quality, accessibility, and governance of an organization’s data assets are foundational to any AI initiative. Models are only as good as the data they are trained on, and enterprises that lack mature data management practices will struggle to produce reliable AI outputs regardless of how sophisticated their models are.
The report highlights that data readiness is one of the areas where the gap between leaders and laggards is most pronounced. The 13% of mature enterprises tend to have well-established data governance frameworks, clear data ownership structures, and modern database infrastructure that supports both traditional analytics and AI workloads. The remaining 87% often have fragmented data estates, inconsistent quality controls, and limited visibility into how data flows through their organizations. Addressing these foundational issues requires not just technology investment but, again, skilled people who understand both the technical and organizational dimensions of data management.
The Geopolitical Dimension of AI Talent
AI sovereignty is increasingly a geopolitical concern, not just a corporate one. Governments in Europe, Southeast Asia, the Middle East, and Latin America are actively developing national AI strategies that include provisions for building domestic talent pipelines. France’s AI strategy, for example, has allocated significant funding to university research programs and AI-focused graduate education. Singapore’s National AI Strategy 2.0, announced in late 2023, explicitly identifies workforce development as a critical enabler of the country’s AI ambitions.
For enterprises operating in these regions, alignment with national sovereignty objectives is becoming a business imperative. Companies that can demonstrate local AI capability — including locally based talent — are better positioned to win government contracts, comply with emerging regulations, and build trust with customers who are increasingly sensitive to questions about where and how AI decisions are being made. The EDB report suggests that enterprises which treat sovereignty as a comprehensive strategy encompassing people, data, and technology will outperform those that focus narrowly on any single dimension.
What the Next 18 Months Will Determine
The window for enterprises to close the AI readiness gap is narrowing. As generative AI capabilities continue to advance rapidly, the organizations that have built mature, well-governed AI operations will compound their advantages. Those still struggling with foundational issues — talent shortages, data quality problems, unclear governance structures — risk falling further behind in a competitive environment where AI capability is increasingly table stakes.
Industry analysts expect the talent dimension to receive greater executive attention in the second half of 2025, driven in part by high-profile AI failures that can be traced back to human and organizational shortcomings rather than technology deficits. The EDB report serves as a useful corrective to the prevailing narrative that AI transformation is primarily a technology procurement exercise. The data tells a different story: the organizations that will win the AI era are those that invest as aggressively in their people as they do in their platforms.
For enterprise leaders reading the 13% figure, the appropriate response is not despair but honest self-assessment. The path to AI maturity is well understood — it requires integrated investment across technology, data, governance, and talent. The challenge is execution, and execution is, as it has always been, a fundamentally human endeavor.