Budget 2026 unequivocally signals Malaysia’s all-in strategy on Artificial Intelligence, positioning it as a core pillar of our national future. The financial commitments are broad and substantial, spanning a nearly RM5.9 billion allocation for cross-ministry research and development, a RM2 billion Sovereign AI Cloud, and various funds to spur industry training and high-impact projects. This ambition is commendable, but ambition, even when well-funded, is no guarantee of success. The critical question now shifts from “what” to “how,” and it is in the execution where our grand vision will either take flight or falter.

A central pillar of our AI strategy is the National AI Office (NAIO), and its RM20 million allocation is a welcome start. The challenge ahead is not a lack of commitment from our various ministries and agencies, which are already pursuing valuable AI initiatives. Rather, it is the risk of fragmentation. To transform these individual efforts into a powerful, cohesive national programme, NAIO’s role must evolve beyond coordination to strategic command. This does not mean replacing the excellent work being done, but empowering NAIO with a cross-ministry portfolio view to prevent redundancy, harmonize standards, and ensure every ringgit of public funds is maximized. By creating a central registry of government AI projects and a single outcomes framework, we can amplify the impact of each agency’s work, ensuring that parallel efforts are converted into a unified, national success story.

Similarly, the budget’s emphasis on talent development is rightly placed. But training more AI graduates is only half the equation; we must ensure our industries are ready to integrate them effectively. Simply funding courses is not enough. We should consider making training grants conditional on tangible outcomes: verified industry placements for graduates, a focus on open, cross-platform tools to avoid proprietary lock-ins, and requirements for short, in-situ implementation cycles with documented results. This ensures we are building a workforce for the real world, not just for the classroom.

The budget’s focus on sovereignty, marked by the launch of the ILMU language model and the Sovereign AI Cloud, is a laudable inflection point. But true sovereignty is not merely about where data resides; it is about who sets the algorithmic and access rules that govern it. The devil, as always, lies in the details. Who will decide which datasets are hosted? How will compute resources be priced for local firms? And most importantly, what are the adoption mechanisms that will compel ministries and SMEs to actually use it? Without clear answers and a robust adoption strategy, even a sovereign cloud risks becoming an impressive but idle monument—a white elephant of good intentions.

One of the budget’s most prescient moves is tasking MIMOS with deepfake detection. This is not a trivial matter; it is a direct response to a clear and present threat. Over the past three years, authorities have had to request the takedown of over 40,000 pieces of AI-generated disinformation. The shocking case in Kulai, where a student allegedly used AI to create explicit deepfakes of schoolmates, brings this danger into sharp focus. This initiative is a crucial and necessary step towards safeguarding our national security and public safety.

Budget 2026 has laid the financial groundwork. It has signaled our intent to the world. If Malaysia is to truly become an AI nation by 2030, the focus must now pivot from macro announcements to micro-implementation. The next budget must not only allocate for global data centres and grand projects, but for the hard, unglamorous work of driving local AI adoption across our SMEs and public services. That is the true measure of a national programme.