Singapore AI Adoption 2026 — The MOM Survey Decoded for CEOs

Analysis by Prabjeet Singh Anand. CEO, PeopleCentral. Founder, AI CEO Brief. 20 years building businesses in Singapore. This analysis draws exclusively on the Ministry of Manpower's April 2026 primary source report. All numbers verified against the primary source. Primary source referenced at the end.

In One Minute

Singapore's first firm-level AI adoption study surveyed 2,560 establishments employing 486,600 workers between January and March 2026. The headline numbers reset what most CEOs assume about where Singapore actually sits.

• 28.5% of firms have adopted AI, and only 3.8% have integrated it into core processes
• Size predicts adoption more than any other variable — 3x gap between the smallest and largest cohorts
• The sector gap inside Singapore is larger than the gap between most countries — Information & Communications at 74.1%, Food & Beverage at 21.9%
• Only 6.2% of adopting firms reduced headcount, while 18.9% redesigned roles and 13.9% created new AI-related jobs
• Strategy is the binding constraint that most CEOs underestimate — 32.3% of firms cite it as a barrier, and it blocks every other barrier from being solved

The Singapore AI adoption story is not what the policy headlines suggest. This decoded read explains the gap between policy ambition and firm-level execution, and what CEOs should do about it.

1. The Headline Number, and Why It Matters

71.5% of Singapore firms have not adopted AI yet.

The Ministry of Manpower's April 2026 report is Singapore's first firm-level AI adoption study, surveying 2,560 private sector establishments employing 486,600 workers. Singapore is one of the most digitally competitive economies on earth, and seven in ten of its companies have not started.

That is the gap that matters.

Among the 28.5% that have started, the picture is even more sobering. Only 3.8% have integrated AI into core business processes. 7.4% are planning to integrate. 6.0% are piloting. The remaining 11.3% are still exploring or evaluating.

That last number is the one most coverage misses. Awareness of AI is everywhere in Singapore. Boardroom conversations about AI happen weekly. Deployment at scale is rare.

The MOM report sets a real baseline against which the next two years of Singapore policy — including the Enterprise Innovation Scheme deduction, Champions of AI programme, and the upcoming Enterprise Development Grant for AI (EDGE) — will be measured.

2. Size Predicts Adoption More Than Any Other Variable

The single strongest signal in the MOM data is firm size. AI adoption rises monotonically with employee count.

The gap between the smallest and largest cohorts is approximately 3x. This is the most important pattern in the data for anyone designing AI strategy in Singapore.

The gap is not a confidence gap. SME CEOs are as aware of AI as enterprise CEOs. The gap is a resource gap. Larger firms have dedicated IT teams, governance capacity, and budget for experimentation that smaller firms do not. They can absorb the cost of running multiple pilots, fail on some, and still move forward.

For Singapore policy, this means that the headline adoption rate is not a single number that improves over time. It is at least two numbers — SME adoption and enterprise adoption — moving on different trajectories with different constraints.

For CEOs, the implication is operational. A 50-person company should not benchmark against a 500-person company. The relevant peer set for any business is firms of its size in its sector. The 25-to-99 employee cohort is at 29.2% adoption. A company in that range that has done nothing on AI is not at the back of the pack — it is with the majority — but the gap to close is also smaller than most CEOs assume.

3. The Industry Gap That Should Worry Some CEOs

After size, sector is the next strongest predictor. The gap inside Singapore is larger than the gap between most countries.

Three patterns are worth pulling out.

The top three sectors are racing. Information and Communications, Professional Services, and Financial Services are each above 56% adoption. The competitive dynamic in these sectors has already shifted. A firm in the bottom quartile of its sector is now structurally behind its median, and the gap is widening every quarter.

The bottom three sectors are not yet in the race. Retail Trade, Food and Beverage, and Construction are each below 26%. The strategic question for CEOs in these sectors is not whether to adopt AI. It is whether to be the first mover in their sub-segment. The first F&B chain that runs AI-enabled scheduling, inventory management, and customer service at scale will pull away from the rest of the industry.

The middle six are the most strategically interesting. Manufacturing at 33.4%, Real Estate at 35%, Transport at 32.1%. These are sectors where one or two well-executed AI deployments shift the competitive landscape inside 12 months. The window for those moves is open right now.

4. AI Is Not Killing Jobs in Singapore — Yet

The most counter-narrative finding in the MOM data is the workforce impact. Among the 28.5% of firms that have adopted AI, the impact on headcount is the opposite of what most coverage suggests.

This is not the same story as the headlines from the United States. Singapore is following a different model. AI is taking tasks. Jobs are being redesigned around the AI.

For APAC CEOs, this matters because most AI workforce strategy being written in 2026 is still shaped by the US narrative. The US narrative is "AI is taking jobs." The Singapore data tells a more useful story: AI is taking tasks, and the companies that thrive are the ones that redesign roles around it.

The risk is complacency. The MOM data captures a snapshot from early 2026. The reductions that have not yet happened can still happen. The model holds only if CEOs deliberately design for augmentation rather than drift into elimination.

5. Where Singapore Sits Globally

Singapore's 28.5% firm-level adoption rate places it in an unusual position. Higher than most of APAC. Lower than the most digitally advanced European economies.

The interesting comparison is not the rank order. It is the gap between policy ambition and firm-level execution.

Singapore has the strongest AI policy framework in APAC — the National AI Strategy 2.0, the Enterprise Innovation Scheme deduction, the Champions of AI programme, the MAS-led financial sector guidance, the upcoming EDGE programme for SME funding. By any measure of policy effort, Singapore is at the front of the pack globally.

The output — actual firm-level deployment — lags Denmark and Finland by 12 to 18 percentage points. The policy is doing its job. Companies are not yet using what the policy makes available.

6. What Is Actually Blocking Adoption

The MOM report ranks the barriers to AI adoption across Singapore firms.

The reflex is to solve the top-cited barrier first. That is the wrong move.

Cost is the most visible barrier and the easiest to articulate at board level. It is also the most fixable with budget and the most directly addressed by Singapore policy. Cost as a barrier is real, but it is not the binding constraint for most firms.

Lack of clear strategy — 32.3% overall, 32.4% among smaller firms. This is the barrier that quietly stops every other barrier from being solved. If a CEO does not know what they are deploying AI for, no amount of budget fixes it. Strategy is the prerequisite. Most Singapore firms do not have one specific to AI.

The MOM data reveals a second hidden pattern. Larger firms cite very different barriers than smaller firms.

Larger firms (more than 500 employees) — top barriers:

• Integration complexity: 56.1%
• Data security: 55.4%
• Lack of expertise: 53.5%
• Cost: 45.9%

Smaller firms — top barriers:

• Cost: 44.7%
• Expertise: 42.1%
• Integration: 37.2%
• Strategy: 32.4%
• Trust: 30.8%

These are different problems requiring different solutions. For larger firms, the binding constraint is governance. For smaller firms, the binding constraint is direction — knowing what to do and being confident enough to commit resources to it.

7. The Two Singapores

The MOM data tells the story of two different Singapores operating side by side.

Singapore A is the digitally competitive economy of policy headlines. Information and Communications sector at 74.1% AI adoption. Enterprise firms at 76.4%. Professional Services and Financial Services close behind. World-class governance frameworks, internal AI councils, capability building budgets.

Singapore B is the broader economy where most workers and most firms actually sit. F&B and Retail under 25% adoption. SMEs under 30%. Strategy as the binding constraint. Most firms with awareness but no plan.

Both Singapores are real. The policy headlines describe Singapore A. The MOM data describes the full distribution.

The strategic question for a Singapore CEO is which Singapore they are operating in, and which Singapore their customers, suppliers, and competitors are operating in.

8. The CEO Playbook

The MOM data is rich. Most CEOs do not need to memorise the whole report. They need a small number of decisions that the data should change.

For Singapore SME CEOs

One — Do the strategy work first. If your AI strategy fits on one page, you have one. If it does not exist, do that work this quarter before any tool purchase. The one-page document should answer three questions: which workflows are first, which tools are first, which metrics matter.

Two — Benchmark against your size cohort, not the national headline. If you have fewer than 100 employees, your peer set is at 29.2% adoption. Translate that into a 90-day deployment plan.

Three — Use the training portion of available policy. Most SMEs use the Enterprise Innovation Scheme deduction for software but not for training. Training is the higher-leverage spend. The policy supports both.

For Singapore enterprise CEOs

One — Set up the governance layer before deploying. If your AI deployment plan does not include who watches the agent, who has the kill switch, and who is accountable when something drifts, the plan is not finished.

Two — Audit your sector position. If you are in Information and Communications, Professional Services, or Financial Services and not above the sector median for adoption, your competitors have a 12 to 18 month lead. Your strategic question is no longer whether to move. It is how fast.

Three — Design role redesign as a deliberate programme, not a side effect. The 18.9% of firms that redesigned roles after adopting AI did better than the 6.2% who reduced headcount. Redesign is the model. Reduction is the failure mode. Choose deliberately.

For APAC regional CEOs operating into Singapore

One — Singapore is your test market, not your scale market. The policy infrastructure, the regulatory clarity, and the talent density make Singapore the right place to test AI deployments before scaling regionally.

Two — Singapore enterprise clients are now Pioneer-tier in expectation. Financial services and Information and Communications clients in Singapore now expect AI-enabled vendors. SME clients are still Transitional. Build vendor offerings that address both segments distinctly.

Three — Singapore is a talent platform for AI roles regionally. As Singapore firms train workers in AI fluency, the regional talent market shifts. Build hiring strategy around this.

Frequently Asked Questions

What does MOM stand for?

Ministry of Manpower. The Singapore government ministry responsible for labour market policy. The April 2026 report is the first time MOM has surveyed firm-level AI adoption across the private sector.

How was the MOM survey conducted?

The MOM Manpower Research and Statistics Department surveyed 2,560 private sector establishments employing 486,600 workers between January and March 2026. The sample was designed to be representative of the Singapore private sector economy.

What does "AI adoption" mean in this study?

MOM defined AI adoption as the use of AI technologies including machine learning, natural language processing, computer vision, and generative AI for business operations. Firms were classified as "adopting" if they were exploring, evaluating, piloting, planning, or integrating AI into business processes.

Is 28.5% adoption good or bad?

It depends on the benchmark. Compared to most APAC economies, Singapore is ahead. Compared to Denmark (47.5%) and Finland (41%), Singapore is behind. Compared to Singapore's own policy ambition and infrastructure quality, the gap suggests an execution problem at firm level rather than a policy problem.

Why is the size gap so large?

Larger firms have dedicated IT teams, governance capacity, and budget for experimentation that smaller firms lack. They can absorb the cost of running multiple AI pilots, fail on some, and still move forward. SMEs face a binary choice — invest meaningfully or not at all.

Which sectors should adopt AI fastest?

Sectors with high cognitive task density adopt fastest — Information and Communications, Professional Services, Financial Services. Sectors with high physical or interpersonal task density adopt slowest — Food and Beverage, Retail, Construction.

Will AI eliminate Singapore jobs?

Based on the MOM data, not at scale yet. Only 6.2% of adopting firms reduced headcount. 18.9% redesigned roles, 13.9% created new AI-related jobs, and 11.2% redeployed workers. The Singapore model is currently augmentation, not elimination. Whether this pattern holds depends on deliberate CEO decisions over the next two years.

What is the biggest barrier to AI adoption in Singapore?

The MOM report identifies cost as the most-cited barrier at 44.9%. But strategy at 32.3% is likely the binding constraint — the barrier that blocks every other barrier from being solved. Cost is fixable with budget. Expertise is fixable with training. Strategy must come first.

How does this report relate to the Enterprise Innovation Scheme deduction?

The Enterprise Innovation Scheme (EIS) deduction is one of several Singapore policy tools designed to lower the cost barrier to AI adoption. The MOM data shows that cost remains the most-cited barrier even with EIS available, suggesting that awareness of and use of the deduction may be lower than expected, or that other barriers compound to keep cost-conscious firms from acting.

Will MOM publish updated data?

MOM has indicated this is intended as a recurring study. Future iterations will allow tracking of adoption trends over time. The next release is expected in 2027.