Artificial intelligence has become one of the few technologies in recent memory that organizations are simultaneously overestimating and preparing for. Across Asia and the Pacific, levels of investment suggest tremendous confidence in AI’s ability to reshape enterprise performance. The organizational foundations required to realize this value tell an entirely different story.
KPMG’s inaugural Global AI Pulse survey, published in March 2026 and drawing on more than 2,100 senior executives in 20 countries, found that organizations across Asia-Pacific plan to invest an average of US$245 million in AI over the next twelve months. This figure exceeds the Americas ($178 million), EMEA ($157 million) and the global weighted average ($186 million).
The commitment is real, and so is the gap. IDC predicts that by the end of 2026, 45% of AI-powered digital use cases across Asia-Pacific and Japan will fail to meet their return on investment targets. More interestingly, IDC attributes these failures largely to poor data bases and unclear value realization rather than deficiencies in the technology itself.
This distinction matters because it suggests that the region does not have an AI investment problem. There is a problem with running the AI. And they are solved in very different ways. Pursuing AI for its own sake is rarely transformative. More often than not, it creates the appearance of progress without addressing the underlying conditions required to support it.
The instinct when AI initiatives disappoint is to look for technical explanations. The model was inadequate. Implementation was rushed. The seller over promised. The platform failed to scale.
Each of these explanations contains some truth. That said, Grant Thornton’s 2026 AI Impact Survey points to a more fundamental issue. In many organizations, approval came before accountability.
Boards approved investments without setting governance expectations. Leadership teams deployed AI capabilities without clearly defining ownership. Programs were scaled before anyone agreed on how success would be measured beyond the metrics of activity and enthusiasm of the pilot phase.
The result is what the survey describes as a growing “proof gap”. The unfortunate implication is that the gap between AI investments and AI outcomes is not primarily a technology problem and therefore cannot be solved by technology choices alone.
IDC’s data makes this challenge evident at scale. About 75% of organizations across Asia-Pacific and Japan have already deployed agent AI in some form. And 61% of CEOs identify agentive and generative AI as their highest priority investment area. However, more than a third of organizations remain stuck in experiments and isolated point solutions, unable to move to enterprise-scale deployment.
Investments have moved rapidly. Government, on the other hand, seems to have taken a more reflective approach. Those who have spent time leading creative, marketing and manufacturing operations across regulated, matrixed, multi-market environments will recognize the pattern immediately.
Organizations that generate meaningful returns from AI are rarely the first or fastest movers. They are not necessarily the ones with the most sophisticated tools. Most often, they are the organizations that have created clarity before the scale.
They defined success at the workflow level rather than the strategy slide level. They built measuring frames before widespread deployment. They invested in the discipline of data before they invested in the ambition of AI. And they set ownership before they set budgets.
Now, none of these tend to make keynote presentations. What it is, however, is extremely effective. This is why conversations about AI increasingly need to shift from capability to operational models.
The real question is no longer whether organizations can access AI – most already do. The most important question is whether they can coherently integrate AI into the way decisions are made, work is executed and value is measured. This is a fundamentally different challenge.
Gartner’s forecast of $2.59 trillion in global AI spending by 2026, representing 47% year-over-year growth, comes just as the technology is firmly entrenched in the so-called Valley of Disappointment. The comparison is revealing.
Capital continues to flow aggressively into AI as many organizations still struggle to demonstrate value at scale. The Asia-Pacific now finds itself at the center of this tension. The opportunity of the region is important. IDC projects that by 2030, half of all new economic value generated by digital businesses in Asia-Pacific will come from organizations investing in AI today.
But this future will not be determined by investment levels alone. It will be determined by execution. Organizations that succeed will not necessarily be those that spend the most or move the fastest.
Enterprise history is littered with examples of organizations mistaking early adoption for sustainable advantage. They will be the ones who treat data governance, measurement, accountability and discipline as prerequisites rather than afterthoughts.
They will also admit that the first wave of productivity gains is simply the entry fee. Competitive advantage emerges when AI reframes the way work is organized, measured and scaled.
Organizations that see real impact treat AI as an operating model shift rather than a technology deployment, redesigning how humans and machines work together rather than removing expert judgment from the loop.
The matter lies beyond the legal operations in which he created it. The first wave of productivity gains is simply the entrance fee. Once automation becomes mainstream, the focus shifts to the last mile, the point at which customer value is actually created and how well an organization orchestrates everything around it.
The gap between AI investments and AI outcomes is not a reason to moderate ambition. It’s a reminder that ambition, without the organizational architecture to support it, rarely scales.
Amardeep Devadason is global head of marketing at Williams Lea, an RRD company, and head of growth for creative and marketing solutions at RRD. He drives global growth and marketing agendas across regions, with a focus on enterprise-scale AI-augmented creative manufacturing and marketing.
