UAE organisations increased artificial intelligence spending by 105% year on year, but execution has not kept pace, with the country recording an AI maturity score of 48 out of 100, according to UAE-specific findings from ServiceNow’s Enterprise AI Maturity Index.
The score improved by 13 points from the previous year, indicating progress, but the study suggests many organisations are still struggling to move from AI experimentation to enterprise-wide execution.
ServiceNow said the gap is being driven less by insufficient investment and more by fragmented technology environments, disconnected data, siloed workflows and limited maturity in AI-enabled operations and workforce development.
AI spending rises faster than operational maturity
UAE organisations are increasing investment aggressively, with AI expected to account for almost one-fifth of total IT budgets by 2027.
Despite that spending, many businesses are attempting to layer AI onto legacy systems and fragmented operating environments.
The study found stronger progress in AI vision, strategy and leadership than in execution.
That imbalance suggests enterprises are increasingly clear about what they want AI to achieve, but are still struggling to build the technical and organisational foundations required to scale it effectively.
Agentic AI adoption remains cautious
Agentic AI has become one of the region’s most prominent technology themes, but the research suggests UAE organisations are still taking a cautious approach to autonomous deployment.
Around 57% of organisations have implemented agentic AI in some form, but only 7% have used it to build autonomous workflows.
In most cases, AI is still being used to assist employees rather than independently orchestrate end-to-end business processes.
Saif Mashat, vice president for the Middle East and Africa at ServiceNow, said the organisations making the strongest progress are moving beyond isolated pilots toward AI orchestration that connects legacy systems, data, governance and AI agents within a common control layer.
Legacy systems remain a major constraint
Only 14% of UAE organisations surveyed have replaced legacy systems with integrated platforms.
That means AI is often being deployed across disconnected applications and workflows rather than a unified operational backbone.
This limits the ability of organisations to automate processes consistently and makes it harder to scale AI across different business functions.
The findings suggest infrastructure modernisation remains one of the most important prerequisites for moving from isolated AI use cases to enterprise-wide transformation.
Data readiness is another major barrier
Data quality and accessibility also remain significant challenges.
According to the study, 77% of UAE executives cite inadequate data accuracy, access and management as a major barrier to AI adoption.
This reinforces the importance of modernising data infrastructure alongside AI deployment.
Without reliable and integrated data, organisations may struggle to build trustworthy models, automate decisions or use AI consistently across workflows.
Governance capabilities remain underdeveloped
The report also highlights a gap in AI governance.
Only 16% of UAE organisations have implemented AI testing, auditing and risk management processes.
That suggests many enterprises are still developing the controls required to scale AI safely.
The gap becomes more important as organisations move toward agentic systems that can take actions autonomously and interact with multiple business systems.
Higher-maturity organisations report stronger returns
ServiceNow said organisations with the highest AI maturity levels take a broader approach that combines technology, data, governance, workforce development and autonomous workflows.
According to the report, these organisations achieve an average AI return on investment of 160%, with that figure projected to rise to 194% within two years.
They are also reported to be 5.6 times more productive, 2.7 times more successful at scaling AI and 2.6 times more effective at managing risk.
The source does not provide the underlying methodology for calculating those performance differentials.
Why this matters
The findings suggest the UAE’s AI challenge is shifting from investment to execution.
Enterprises have already committed significant budgets and strategic attention to AI, but the ability to generate returns depends on whether they can integrate systems, improve data quality, build governance and redesign workflows around the technology.
That means the next stage of enterprise AI maturity is likely to be less about adding more tools and more about creating the operating environment in which AI can work at scale.
Editor’s note
The most important takeaway is that AI maturity is becoming an orchestration problem.
UAE organisations are spending heavily, but fragmented systems, weak data foundations and limited governance are preventing that investment from translating into broad operational change.
The gap between 57% adoption of agentic AI and just 7% deployment of autonomous workflows captures the issue clearly. Many enterprises have access to the technology, but far fewer have built the foundations required to let it operate safely and independently across the business.
The organisations that close that gap fastest are likely to be those that treat AI transformation as an enterprise architecture challenge rather than a software procurement exercise.
