- Malaysia’s enterprise AI readiness gap stems from fragmented data.
- AI deployment depends more on governed data than model choice.
Malaysia’s plan to become a regional data and AI hub by 2030 is raising questions about whether enterprises have the data systems needed to deploy AI beyond pilots.
Cecily Ng, Vice President and General Manager of ASEAN and Greater China at Databricks, said Malaysia has shown strong AI ambition at both national and enterprise levels. The country is positioning AI as a core economic driver, with efforts covering adoption, governance and talent.
Ng said organisations in sectors, including Malaysia Airlines, PETRONAS, and Digital Nasional Berhad, are using Databricks to bring AI into workflows and support data-driven insights.
Data foundations remain the main gap
The main gap between national ambition and enterprise-level readiness remains the data foundation, according to Ng. Many organisations still operate in multiple clouds and legacy systems. Data also remains separated in business units.
“What we are seeing in conversations with Malaysian business leaders is that most organisations are still dealing with fragmented data estates,” she said. “That makes it very difficult to operationalise AI beyond isolated use cases.”
Leadership teams are discussing AI at a level, but the underlying data layer is often still being unified and governed before it can support wider deployment.
Ng cited Malaysia Airlines as an example. The airline consolidated data from internal and third-party environments onto a single platform to support near real-time insights and AI-assisted customer segmentation.
Moving AI beyond pilots
Organisations that move beyond pilots tend to treat AI as part of a wider data and business transformation, not a standalone innovation project, Ng said. Many companies have tested generative AI through chatbots or productivity tools. Moving those tools into production requires stronger data systems.
“In 2026, the organisations succeeding with AI are not simply testing more models or building more demos,” she said. “They are investing in the foundations that allow AI to be trusted and scaled in the business.”
Companies moving into production are focus on use cases with clear returns. These often begin in internal functions like operations and risk. Such deployments allow organisations to prove value and strengthen governance before expanding AI into customer-facing services.
Ng said organisations that see value from AI tend to align business and technology teams around a small number of use cases before scaling them.
Ng also cited Digital Nasional Berhad, which worked with Databricks on a unified data and AI foundation for network and operational workflows. The platform supports real-time network data processing and performance monitoring. It also supports anomaly detection and decision optimisation, contributing to up to 70% cost savings.
Models are not the only issue
Enterprises risk treating AI too narrowly as a model selection issue, according to Ng. Foundation models are attracting attention, but they are becoming more available, and no single model performs best in every task.
“The real differentiator is an organisation’s data – how well it is unified and made available for AI workloads,” she said.
Ng said proprietary enterprise data is also central to production AI. Public models are available, but a company’s own customer interactions, operations, product use, and domain knowledge help make AI outputs more relevant to the business.
“From a Databricks perspective, the future is model-agnostic and multi-model,” Ng said. “Customers should be able to use the best model for each task, while keeping enterprise data secure and observable.”
Responsible AI needs operational controls
Responsible AI needs to be embedded into the AI lifecycle, not handled only through policies or ethics statements, Ng said. This is especially important in regulated and operationally complex sectors like aviation, energy and telecommunications.
Organisations need visibility into how AI systems make decisions. That includes the data they access, the tools they use, and the way outputs are generated. This is becoming more important as AI systems become more agentic and autonomous. “Responsible AI at scale is fundamentally about trust and governance being built directly into the AI lifecycle,” Ng said.
Ng also cited AIA for example outside Malaysia. The insurer brought internal customer data and external behavioural analytics data into one secure Databricks platform, allowing financial advisers to provide personalised recommendations. That led to a doubling of customer engagement and lead generation, according to Ng.
Malaysia’s regional position
Across ASEAN and Greater China, markets that are further ahead in enterprise AI have moved quickly to connect data strategy with business outcomes, Ng said. These organisations have invested in unified platforms and embedded AI into operations like risk management, supply chain optimisation, and customer engagement.
Malaysia’s cloud-forward organisations and infrastructure investment give enterprises a chance to build unified data and AI architectures earlier, instead of addressing fragmented systems later, according to Ng.
Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is part of TechEx and is co-located with other leading technology events, click here for more information.
Tech Wire Asia is powered by TechForge Media . Explore other upcoming enterprise technology events and webinars here .