Originally published 10 March 2026. Updated to reflect confirmed details from Budget 2026 and subsequent government announcements.
Singapore has established a National AI Council, chaired by Prime Minister Lawrence Wong. On the surface, this sounds like another government committee. It is not. A PM-chaired council with a mandate to define clear rules for how AI is developed and used signals a materially different approach from a ministry advisory group or a voluntary framework consultation process.
This article explains what the Council signals, what it has confirmed it will do, and what mid-market companies in Singapore should be doing right now.
Why a PM-chaired council matters differently
Singapore has had AI governance frameworks and advisory bodies for several years. IMDA published the first version of its Model AI Governance Framework in 2019. MAS has issued AI risk guidance for financial services. The Smart Nation initiative has been running AI governance programmes since the early 2020s.
What has been missing is a body with clear authority to set rules rather than recommendations, positioned at the highest level of government, and accountable for the outcomes of AI development across the economy.
The National AI Council fills that gap. When the Prime Minister chairs a body himself, it signals two things. First, that AI governance is now a strategic national priority, not a regulatory compliance matter managed by individual agencies. Second, that the Council’s outputs will carry the weight of national policy, not the weight of a guidance document that businesses can choose to follow or not.
PM Wong’s Budget 2026 speech was explicit: “We will define how AI is developed and used in Singapore. We will set clear rules to ensure it is applied responsibly and safely.” That language is deliberate. The distinction between rules and principles signals an intent to move from guidance toward enforceable requirements.
What the Council has confirmed it will do
Unlike earlier advisory bodies, the National AI Council was announced with a defined mandate. Several elements are already confirmed policy, not inference.
Four sector missions. The Council oversees National AI Missions in advanced manufacturing, connectivity and logistics, finance, and healthcare. Each mission is anchored in clear objectives and tangible outcomes, not aspiration. If your business operates in any of these sectors, the governance expectations that follow will be sector-specific, not generic.
Coordinated regulatory alignment. The Council’s mandate includes aligning R&D, regulation, and investment promotion so that different agencies pull in the same direction. The fragmentation between IMDA, MAS, and PDPA frameworks that has made compliance planning difficult is a known problem the Council is designed to resolve. A consolidated national AI regulatory framework is the logical output.
Regulatory sandboxes. PM Wong confirmed the government will review regulations and create sandboxes so companies can test AI innovations safely. This suggests a structured pathway for deployment in regulated sectors that does not yet exist.
The Champions of AI programme. For firms with the ambition to use AI to comprehensively transform their business, the Council is launching a tailored support programme. Access is not automatic. It presupposes governance readiness. Businesses that cannot demonstrate accountable AI deployment are unlikely to qualify.
Significant tax incentives. For Years of Assessment 2027 and 2028, businesses can claim a 400% tax deduction on AI-related expenditures under the Enterprise Innovation Scheme, capped at S$50,000 per Year of Assessment. The Productivity Solutions Grant has also been expanded to cover a wider range of AI-enabled tools. If your organisation is making AI investments and not structuring them to qualify under these schemes, that is a conversation worth having with your finance team now.
The three governance questions every business should answer now
Before the Council’s first major regulatory publication, every Singapore business with AI in production should be able to answer three questions clearly.
What AI systems are you running and how have you classified their risk? This is the baseline. If you do not have an inventory of your AI deployments with a documented risk classification for each, you are not in a position to respond to governance requirements, because you do not know what you have.
Who is accountable for AI governance in your organisation? Not who is responsible for building AI systems. Who is accountable for the governance of those systems. There is a difference, and most organisations do not have a clear answer.
What would you show a regulator if they asked to see your governance evidence today? Not what documents you could create quickly if you knew a regulator was coming. What you would show them today. If the honest answer is “not much”, that is the gap the Council’s framework will formalise into an obligation.
A sharper point for regulated sectors
If your business operates in finance or healthcare, the urgency is higher than the general picture suggests. These sectors are not just adjacent to the Council’s mission priorities, they are the explicit focus. Sector-specific governance requirements are the most probable near-term output, and the financial services sector already has MAS AI risk guidance as a precursor.
Governance gaps that feel manageable today will become enforceable obligations more quickly in these sectors than elsewhere. That is not speculation; it follows directly from where the Council has chosen to focus.
What to do now
Singapore’s technology regulation has followed a consistent pattern: voluntary guidance becomes enforceable obligation over a relatively short window. The PDPA, MAS’s technology risk frameworks, and IMDA’s sector guidance all followed this trajectory. The National AI Council is operating at a higher level of government authority than any of those predecessors.
Businesses that establish their governance baseline now will find the transition significantly easier than those that wait. The most important first step is understanding where you actually stand. Not where you think you stand, and not where you would stand if you had a few months to prepare. Where you stand today.
The AI Governance Review is a starting point for that: a structured conversation to identify where your current posture is strongest, where the most significant gaps are, and what a realistic plan to address them looks like.
For smaller businesses, the SME Governance Starter is designed as the right entry point: structured, affordable, and scoped to what a mid-market company actually needs to demonstrate credible governance without the overhead of an enterprise programme.
If you have AI in production in Singapore, this is the right time to make that call.
Arjen Hendrikse is the founder of Aivance Consulting and an ISO/IEC 42001:2023 Lead Auditor. He tracks AI policy and regulation across Southeast Asia and helps mid-market companies and enterprises build governance frameworks that hold up to regulatory scrutiny.