AI may have dominated the conversation at Boomi World Forum Singapore 2026, but one message came through clearly – for many organisations, the biggest barrier to AI adoption is the infrastructure beneath it.
That challenge was laid out in AI Ambition Meets Data Reality: APAC Technology Priorities and Challenges 2026, research conducted by Omdia and commissioned by Boomi, covering more than 1,100 senior technology and business decision makers across Australia, New Zealand, Singapore, Malaysia and the Philippines.
The numbers show an APAC market moving rapidly from AI experimentation to deployment. Some 74% of organisations already have AI initiatives underway, while 95-percent have committed funding to AI. Nearly half already have AI in production, and 93-percent expect AI-enabled automation to significantly affect business processes within the next two to three years.
The integration gap behind the AI boom
The challenge is that enterprise architecture is not always moving at the same speed.
While 90-percent of organisations want to move towards a unified, AI-ready integration platform, only 46-percent currently have one in place.
That gap is worth clicking into because enterprise AI increasingly depends on information flowing between cloud platforms, databases, SaaS applications, APIs and operational systems. If those environments remain fragmented, AI systems risk operating without complete, trusted or up-to-date enterprise context.
The issue becomes even more significant as AI agents begin interacting directly with business systems.
According to the study, 81-percent of organisations say unmanaged or “shadow” integrations are already affecting data quality and confidence. For IT teams, shadow integration is nothing new, but AI raises the stakes. Poor data feeding a dashboard may result in a bad report. Poor data feeding an autonomous agent can translate directly into bad decisions or actions.
Governance is moving slower than deployment
Governance presents a similar challenge.
Only half of the organisations surveyed currently have AI-specific data policies in place, while 22-percent have no defined KPIs or metrics for measuring whether their AI initiatives are succeeding. At the same time, 93-percent recognise that AI will make data quality and governance increasingly important.
That distinction between deploying AI and operationalising it is becoming critical.
Enterprises can introduce co-pilots, generative AI tools and agents relatively quickly. Turning those investments into sustainable business value requires much stronger discipline around data quality, access, governance, measurement and accountability.
There are signs that organisations are responding. Around 89-percent of respondents said they are working to reduce technology sprawl, while 92-percent are consolidating technologies across areas including integration, APIs and automation.
Agentic AI raises the stakes
That consolidation will become increasingly important as agentic AI moves further into the enterprise.
Unlike a standalone chatbot, an AI agent performing real business work needs access to systems, data and workflows. As organisations give these agents greater autonomy, questions around permissions, integration, observability, governance, and security become just as important as the capabilities of the AI model itself.
Data sovereignty is also entering the conversation. The research found that 76-percent of APAC organisations are concerned about sovereignty, but only 24-percent have taken meaningful action.
For businesses operating across multiple APAC markets, where regulatory and data residency requirements can differ significantly, this adds another layer of complexity to enterprise AI architecture.
AI readiness is becoming a data question
The takeaway from Boomi World Forum Singapore was clear: the enterprise AI race is no longer just about who has the most powerful model.
It is about who can connect AI to the right data, systems and workflows – and do it securely, reliably and at scale.
APAC organisations have no shortage of AI ambition. What will separate the pilots from the real transformation is whether the enterprise underneath can keep up.
