Artificial intelligence has entered a new phase.
The initial excitement surrounding generative AI has largely given way to a more practical conversation about implementation. Executive teams are no longer debating whether AI will influence their industry. Most have accepted that it will. The focus has shifted toward understanding how AI can be deployed responsibly, where it can create measurable value, and how organizations can move beyond experimentation into repeatable business outcomes.
Against this backdrop, Gartner's recent warning regarding semantics and AI agents deserves careful attention.
In May 2026, Gartner highlighted what it believes is a growing risk facing organizations pursuing agentic AI. According to Gartner, a lack of semantics is contributing to inaccurate AI agents, unnecessary complexity and wasted spending. More significantly, Gartner predicts that organizations that prioritize semantics within AI-ready data could improve agentic AI accuracy by up to 80 percent while reducing costs by as much as 60 percent by 2027. The full Gartner announcement can be accessed here.
At first glance, semantics may appear to be a niche technical issue. It lacks the visibility of large language models, AI agents or autonomous workflows. Yet Gartner's research points to something far more significant than a technology challenge. It raises a fundamental question about whether organizations have established enough shared understanding for AI to operate effectively within the reality of their business.
This distinction is important because much of the current AI conversation assumes that access to information naturally creates intelligence. Gartner's findings suggest otherwise. Access alone may not be the limiting factor. Understanding may be.
Most organizations already possess vast quantities of information. Customer data, operational data, financial data, transactional data and market data flow continuously through enterprise systems. Yet when leadership teams ask seemingly straightforward questions, they often discover that different parts of the organization interpret the same information differently. Metrics that appear consistent at an executive level frequently have multiple definitions beneath the surface. Revenue may be calculated differently across functions. Customer classifications may vary across business units. Operational measures may not align with financial measures. What appears to be a single version of the truth often turns out to be a collection of closely related interpretations.
Historically, organizations have managed this complexity because people are remarkably good at providing context. Experience fills gaps that systems cannot. Employees understand which reports are trusted, which definitions are accepted and which exceptions apply in particular situations. They navigate ambiguity through institutional knowledge accumulated over years of working within the business.
AI changes that equation.
An AI agent has no understanding of organizational history unless that understanding has been made explicit. It does not know why one report is preferred over another. It does not understand the nuance behind a business definition unless that nuance has been formally captured. It cannot infer context that has never been articulated. It operates entirely within the boundaries of the information and meaning provided to it.
This is where Gartner's emphasis on semantics becomes particularly relevant.
Semantics is often described as the study of meaning, but in a business context it is more practical than that. It represents the shared understanding that allows people, processes and systems to interpret information consistently. It connects data to business reality. It provides the context required to understand not only what information exists, but how that information should be interpreted, governed and applied.
Without that context, AI agents may become highly efficient at producing answers that appear credible while remaining disconnected from the way the business actually operates.
This challenge becomes increasingly significant as organizations move from AI-assisted work toward AI-influenced decisions. The value of AI is not determined by its ability to retrieve information. The value emerges when organizations trust the outputs enough to act upon them. Trust, however, is rarely created by technology alone. Trust emerges when information is understood, ownership is clear, governance is transparent and business meaning is consistently applied.
This is where Gartner's report becomes particularly interesting because it shifts the discussion away from AI capability and toward organizational capability.
Many organizations continue to approach AI as a technology initiative. Gartner's research suggests that the more consequential challenge may be organizational readiness. As AI becomes embedded within business processes, organizations will increasingly need to answer questions that technology alone cannot resolve. Who owns a business definition? Which interpretation of a metric is authoritative? How should governance be applied? Which decisions require human oversight? What constitutes a trusted source of information?
These questions sit at the intersection of business leadership, governance and operating model design. They are not AI questions. They are organizational questions that AI is bringing to the surface.
In many respects, Gartner's findings reinforce a broader shift that is emerging across enterprise data and analytics. For years, organizations have concentrated on creating greater access to information. The next challenge is ensuring that information carries consistent meaning wherever it is consumed. The organizations that succeed in this transition will create an environment where people and AI systems operate from the same understanding of the business rather than competing interpretations of it.
This is one of the reasons I believe business context is becoming an increasingly strategic asset.
Organizations frequently discuss data as an asset. They discuss platforms as assets. They discuss AI capabilities as assets. Yet two organizations can possess similar technology stacks, similar data volumes and access to similar AI models while achieving dramatically different outcomes. The differentiator is often not the technology itself. It is the clarity with which the organization understands its own business.
Organizations that establish shared definitions, trusted ownership models and consistent governance create a foundation that allows decisions to scale. Organizations that fail to establish this clarity often discover that complexity grows alongside technology adoption.
The same principle applies to AI agents.
An AI agent operating within a fragmented environment inherits fragmentation. An AI agent operating within a well-defined business context inherits clarity. The technology may be identical. The outcomes are not.
This is why trusted data products are becoming increasingly important in AI-enabled organizations. Their value extends well beyond making information available. Trusted data products create a repeatable mechanism for connecting information with ownership, governance, lineage, business rules and context. They help ensure that information can be interpreted consistently by both people and machines. In an environment where AI systems are expected to support decisions, that consistency becomes a significant source of competitive advantage.
The most compelling aspect of Gartner's warning is that it reframes the conversation around AI success. The next wave of value creation may not come from deploying more agents, building larger models or introducing additional automation. It may come from establishing enough organizational clarity for those technologies to operate effectively.
That is a fundamentally different challenge.
It requires organizations to think less about what AI can do and more about the conditions required for AI to succeed.
And those conditions are increasingly rooted in business meaning.
What We Are Seeing In North America
“One of the more interesting developments across North America is that AI is creating a new level of executive scrutiny around consistency. Historically, differences in definitions, metrics and reporting approaches could remain largely contained within business functions. AI changes that dynamic because leaders increasingly expect enterprise-wide answers to enterprise-wide questions. As organizations expand the use of AI agents, many are discovering that technology is revealing long-standing alignment challenges that were previously tolerated. The organizations making the strongest progress are treating consistency as a business capability rather than a reporting exercise.”
What We Are Seeing In the UK & EU
“In the UK and Europe, the semantic question is landing with particular force because so many organizations operate across multiple languages, jurisdictions and legacy systems built over decades. When an AI agent is asked what 'customer', 'revenue' or 'exposure' means, the honest answer in many enterprises is that it depends on which country, which business unit and which reporting cycle is asking. Executives are recognizing that this ambiguity, which the organization has quietly absorbed for years, becomes a material risk the moment an autonomous agent acts on it. What we are seeing across UK and EU boardrooms is a renewed focus on shared business meaning as a prerequisite for AI agents — not a nice-to-have. The organizations moving fastest are the ones investing in a common semantic layer their people, regulators and AI agents can all read the same way.”
What We Are Seeing In APJ
“Across APJ, the discussion around AI increasingly centers on scale. Many organizations operate across diverse regulatory environments, languages, cultures and business structures. In that context, preserving meaning becomes as important as preserving data itself. What Gartner highlights through its focus on semantics is particularly relevant for the region because AI must be capable of operating consistently across complexity. Organizations that establish common business understanding across markets will be significantly better positioned to scale AI confidently and responsibly.”
Partner & Customer Perspective
“One theme we consistently observe through partners and customers is that successful AI initiatives rarely begin with technology discussions. They begin with conversations about ownership, accountability and alignment. As organizations prepare for AI, they often discover that different parts of the business hold different interpretations of the same concepts. The most successful programs treat these discoveries as opportunities to strengthen operating models rather than obstacles to implementation. In many cases, the greatest value emerges not from the AI itself but from the organizational clarity created along the way.”
Lili Marsh
Head of Partner & Customer Success
Gartner's warning ultimately shifts the conversation back to where it belongs.
Not on the capabilities of AI, but on the capabilities of the organization deploying it.
Organizations do not become AI-ready simply because they introduce AI agents into existing processes. They become AI-ready when they establish sufficient trust, context and shared understanding for those agents to operate effectively within the realities of the business.
The next phase of competitive advantage may therefore have less to do with artificial intelligence itself and more to do with an organization's ability to articulate what its information means, how it should be interpreted and how it connects to decisions.
As AI becomes increasingly embedded into business operations, that distinction may prove far more important than many organizations currently realize.
The future will not belong to the organizations with the most AI.
It will belong to the organizations that understand themselves well enough for AI to create value.