Across the Asia Pacific, organisations are embracing Generative AI at pace. A new study from Informatica by Salesforce, CDO Insights 2026, shows that two-thirds of APAC companies (66%) have already embedded AI into business processes – an accelerated adoption curve that marks an encouraging milestone.
But it also raises a clear and immediate concern: many of those initiatives are running ahead of the data management, governance and skills foundations required to scale AI safely and effectively.
When AI is treated as a feature rather than a capability, pilots proliferate. Unfortunately, pilots don’t deliver sustained value unless the underlying data is reliable, the workforce is competent in AI use and governance is consistent and robust.
The CDO study reveals that APAC organisations face a significant challenge when advancing AI initiatives beyond experimentation. Nearly nine in ten respondents (89%) identify data reliability as a barrier to moving AI projects from pilot to production. That single statistic explains why so many promising projects stall.
There is, however, an interesting paradox in the region’s mindset. Sixty-seven percent of APAC data leaders report that most or nearly all employees trust the data used for AI. Yet 88% of data leaders are worried that new AI pilots are being launched without first addressing data reliability weaknesses uncovered in earlier projects. In other words, confidence and caution coexist within the same organisation.
This trust paradox arises as employees place blind trust in AI, while CDOs recognise that true AI readiness requires reskilling and governance guardrails. That gap between perceived trust and operational readiness is exactly where risk accumulates.
The CDO study reveals a critical gap: 72% of organisations need better data and AI literacy. While employees often trust data blindly, CDOs know stronger fluency and governance are essential for reliable, responsible AI use. Building AI models is just the start – integrating them into decisions and interpreting results requires deep data and domain expertise. Without this critical combination, organisations risk developing fragile systems that are not only difficult to maintain but also challenging to govern.
In APAC, governance is fragmented – 44% of the organisations surveyed are extending existing data governance frameworks to cover AI, 35% are adopting AI governance tools, and one in five (22%) are building new governance programmes from the ground up.
This fragmented landscape makes it difficult to set region-wide standards for accountability, transparency and auditability – all which regulators and customers increasingly demand.
There is growing recognition of the critical role that data management plays in driving business value. In fact, 86% of APAC organisations plan to increase their investments in data management this year. Their priorities are pragmatic: navigating evolving regulation (43%), strengthening data privacy and security (42%) and improving data and AI governance (39%).
This strategic roadmap is sound, but success depends on translating intent into actions such as improving data quality, unified metadata and lineage, role-based accountability and scale training that elevates foundational skills across the organisation. So, what should organisations in APAC do differently and more urgently to turn AI’s promise into lasting advantage?
First, treat data reliability and completeness as a core product, not just an afterthought. Move beyond ad-hoc fixes by investing in repeatable, measurable data quality processes such as automated profiling, continuous validation, lineage tracking and clear service level agreements for data pipelines to power AI models. When data is managed as a core product, business teams can align on ownership and set clear expectations, driving greater trust and effectiveness in AI initiatives.
Organisations should also adopt practical, risk mitigation-focused governance that supports their AI initiatives by tailoring and applying appropriate controls to different use cases – from simple operational efficiency to customer predictive models. Integrating these measures smoothly into existing processes helps ensure compliance, legal liability; avoid absolutes and keeps AI projects moving forward to scale efficiently and responsibly.
Closing the workforce skills gap is equally critical for realising the full potential of AI. Organisations must deliver targeted, role-specific trainings – building data and AI literacy for business users, empowering engineers with the skills to validate and monitor AI models and equipping leaders with clear accountability frameworks to guide responsible AI deployment decisions.
Finally, maintain transparency with stakeholders – regulators, customers and partners increasingly demand clarity on how AI is governed and applied. Organisations that proactively document decision-making processes, ensure explainability where necessary, and clearly define accountability will build and preserve trust, enabling autonomous agents to function reliably as AI becomes integral to business operations.
Asia Pacific’s rapid AI adoption presents both an opportunity and a responsibility. To harness AI’s full potential, organisations must place robust governance at the forefront, ensuring integrity and reliability of both data and AI-driven outcomes.
Equally important is investing in workforce upskilling to promote the ethical and effective use of these technologies. As in-house capabilities grow and tools and platforms consolidate to lower TCO, businesses can unlock enduring value from their AI investments.
For today’s leaders, the choice is clear: establishing strong data foundations is not just beneficial but essential to advancing AI initiatives beyond experimental phases and achieving scalable, sustainable impact.
This thought leadership piece is written by Richard Scott, Senior Vice President, Asia Pacific & Japan, Informatica from Salesforce.
