For years, Singapore has spoken about artificial intelligence (AI) as a strategic advantage. In 2026, it moved decisively from words to action.
Singapore Budget 2026 marked a clear step‑up from AI ambition to execution. A new National AI Council, chaired by Prime Minister Lawrence Wong, was announced to set direction and oversee national AI missions across sectors. These are not abstract aspirations; they are targeted efforts to embed AI into the economy.
At the same time, the National AI Impact Programme aims to support 10,000 enterprises over the next three years, while the expanded Enterprise Innovation Scheme lowers the cost of AI investment.
Singapore is no longer asking companies whether they want to experiment with AI. It is actively helping them deploy it at scale. As adoption accelerates, responsibility shifts. When AI becomes embedded across operations and decision‑making, control and governance move squarely onto enterprises. As a result, AI sovereignty, which was once discussed mainly by governments, is now a boardroom issue.
Avanade’s Sovereign AI research reflects this shift. Sixty-one per cent of organisations globally say geopolitical events have increased the urgency of adopting sovereign technologies. The reason is simple. AI systems today shape decisions, assess risk and influence outcomes at scale. When these systems operate beyond an organisation’s visibility or control, risk compounds quickly.
The sovereignty gap hiding inside enterprise AI
See also: New playbook to help Singapore enterprises match AI readiness with support
Many organisations believe they are already protecting their digital sovereignty. In reality, most are securing only part of it. While 60% of organisations begin their sovereignty journey by securing data, only 22% extend governance and protection to AI models themselves.
This creates a critical gap.
Data can be stored securely, yet the intelligence interpreting that data, the models generating insights and decisions, often remains poorly governed. If sovereignty stops at the data layer, organisations leave the decision‑making layer exposed.
As AI moves deeper into core business processes, this gap is no longer theoretical. It is where operational, regulatory and reputational risks accumulate. In Singapore’s regulatory environment, where trust, governance and resilience are foundational, such blind spots are increasingly difficult to justify.
When AI starts acting, not just predicting
The challenge is further intensified with the rise of agentic AI. Unlike traditional systems, agentic AI can reason, learn and act autonomously. It can trigger processes, interact with other systems and cross organisational boundaries.
Recognising this shift, Singapore launched the world’s first Model AI Governance Framework for Agentic AI in January 2026. The framework makes one principle clear: even when AI acts autonomously, humans remain accountable.
This signals a broader change in how sovereignty must be understood. It can no longer be defined only by physical control - where servers sit or where data is stored. It must also extend to how AI behaves, who oversees it, and whether its reasoning aligns with organisational values.
Sovereignty now requires three forms of control:
- Behavioural control: defining what AI systems are permitted to do
- Operational control: ensuring humans can monitor, intervene and audit decisions
- Cognitive control: aligning AI reasoning with the organisation’s values, policy and local regulatory requirements.
To stay ahead of the latest tech trends, click here for DigitalEdge Section
AI sovereignty can no longer be defined only by where technology sits, but by how it behaves once deployed. This is why sovereignty can no longer be treated as just a compliance exercise.
Addressing the problem: from compliance to capability
Enterprises need to realise that protecting AI sovereignty does not mean slowing innovation. In fact, the opposite is true. The organisations that scale AI fastest will be those that embed governance into their systems from the start.
Here are practical steps that organisations can look at to get started
- Conduct sovereignty audits for each AI use case: Assess country risk, regulatory exposure and business criticality to determine the appropriate level of control.
- Benchmark infrastructure and services: Organisations must understand where data resides, which cloud services are in use, and which external models or platforms are embedded across their stack.
- Identify systemic risks: AI systems often rely on interconnected applications and providers. Without this exercise, hidden dependencies can surface only during disruption, when it is already too late.
- Establish dynamic governance across the AI stack: Controls must extend across data, models, cloud infrastructure and runtime environments. Static compliance checks are no longer sufficient.
- Invest in real‑time and predictive oversight: AI systems must be continuously monitored to detect anomalies such as prompt injection, hidden backdoors or malicious behaviour.
When governance operates across layers, data, models and compute, enterprises can innovate with confidence while remaining in control.
A leadership moment for Singapore enterprises
Singapore has made its move. AI is now a national execution priority, backed by policy, funding, and clear direction. The next move belongs to enterprises.
AI sovereignty is not about slowing innovation or retreating behind borders. It is about ensuring that as AI becomes more powerful, organisations remain accountable for its actions and outcomes.
The competitive advantage will not come from who adopts AI fastest, but from who governs it best. For Singapore’s business leaders, the time to build that capability is now, before decisions are made by systems they no longer fully control.
Bhavya Kapoor is the president of Avanade Asia Pacific
