By Kirti Pathania, TFGI Insights Contributor
At a glance
- AI adoption is rising among Southeast Asia’s Small Medium Enterprises (SMEs), but meaningful AI integration into business workflows remains limited—risking a widening productivity gap with larger enterprises.
- SMEs need more than technology; they require strong ecosystems, skilled leadership, and targeted financial support to successfully integrate AI.
- By investing in partnerships, capability building, and agile policy interventions, governments can ensure AI strengthens competition, innovation, and inclusive economic growth across ASEAN.
Share this insight
The adoption of Artificial Intelligence (AI) among Small and Medium Enterprises (SMEs) in Singapore tripled in 2024, with businesses using AI across an average of three business functions. While businesses are increasingly adopting AI, a recent survey by Singapore’s Ministry of Manpower (MOM) found that only 3% of enterprises have successfully integrated AI into their workflows.
AI integration goes beyond adoption use cases such as customer service chatbots or employee assistance copilots. It involves embedding contextual use of AI within business workflows. Examples include machine learning (ML) tools that advise supply planners on safety stock planning or procurement agents that may generate automatic requisitions for human approvals.
Although the MOM report indicates that AI integration is still at an early stage, it also highlights significant differences across businesses of different sizes. Early-stage AI adoption rises from 23.9% among firms with fewer than 25 employees to as high as 76.4% among firms with more than 500 employees. These figures illustrate the distinct challenges many SMEs face—including limited strategy, organizational maturity, and resources—which can slow their AI integration journey.
The cost of inaction: Why it matters
SMEs account for more than 97% of businesses across ASEAN. They play a vital role for local economies in the region, contributing between 50% to 85% (across member nations) to domestic employment and 44.8% to regional GDP.
As AI integration advances, there is an opportunity to ensure its benefits are shared broadly. If integration becomes concentrated among larger enterprises with greater financial and organisational capacity, productivity gains may become increasingly concentrated among a smaller group of firms.
Over time, this could create reinforcing advantages for early adopters, enabling them to shape industry standards, pricing models, business processes, and approaches to AI deployment alongside technology providers and consulting firms. Such “success to successful” dynamics may gradually make it more difficult for SMEs to catch up as AI capabilities become embedded across industries. Reinforcing patterns have been observed amongst the Online Travel Agents (OTA) in South East Asia. Larger firms that embraced online platforms and AI technology early were able to vertically integrate their business to adjacent offerings such as hotels, taxis, insurance etc, eventually forc homegrown travel agencies to close, or compelling small businesses to advertise via these platforms and compromise on margins.
If similar dynamics emerge around AI integration, SMEs could be marginalised and face significant barriers to AI integration, eventually leaving them behind.
Levelling the Playing Field for AI Integration
There is an opportunity for the governments and policy makers in Southeast Asia to develop policies which could level the field for AI integration across SMEs and large enterprises.
Broad-based AI adoption alone cannot guarantee inclusive productivity gains. Policymakers must instead tailor interventions to support the diverse ways SMEs adopt and integrate AI throughout their growth journeys.
1. Drive SME AI Integration through Ecosystem Partnerships
SMEs typically adopt AI sporadically to address discrete use cases as their business expands. For instance, a small family-owned food manufacturer in Jakarta recently sought an industry-standard demand planning tool to manage rapid growth in sales, but found enterprise-grade platforms and custom AI consulting roadmaps cost-prohibitive.
Beyond cost barriers, an SME’s AI maturity has a structural dependence on ecosystems, which includes suppliers, regulators, governments, consultancies, investors, and industry partners. Supporting this, the MIT State of AI in Business Report indicates that external partnerships are nearly twice as likely to succeed in integrating AI compared to internal builds. Therefore, policy interventions that respect the non-linear nature of SME adoption—specifically agile initiatives focused on building collaboration—can effectively level the playing field.
Thailand’s SMEs Growth Programme serves as a prime example of an effective ecosystem partnership. This place-based initiative links regional SMEs in the Eastern Economic Corridor and South Thailand with eCommerce and CRM providers. The programme builds a foundation for AI integration while solving three core challenges:
- Skill Gaps: Entrepreneurs gain critical digital skills.
- Cost Constraints: SMEs access trusted technology providers at affordable prices.
- Information Barriers: Businesses learn about relevant technology and software options.
Although the programme helps accelerate digital integration for smaller enterprises, the question remains whether such interventions remain agile and develop further as initial policy results emerge.
While the programme accelerates digital integration for smaller enterprises, policymakers must ensure these initiatives remain agile and evolve alongside emerging results.
2. Invest in Leadership Skills to Strengthen AI Readiness
SME AI maturity is largely influenced by managerial decision-making and leadership’s willingness to experiment with new technologies. For instance, a mid-sized apparel manufacturer in Jakarta seeking to modernise its technology infrastructure faces limited internal IT capacity, restricted budgets for experimentation, and heightened scrutiny over technology investments amid export headwinds from recently introduced United States tariffs.
In many SMEs, the organisational conditions needed for sustained AI integration depend heavily on leadership judgement. Programmes such as Singapore’s SkillsFuture can help strengthen these capabilities by enabling mid-career professionals and senior leaders to access AI training delivered through partnerships with universities and technology providers. These collaborations between government, academia, and technology providers expose participants to emerging developments in AI while equipping business leaders with the knowledge needed to make more informed technology decisions.
According to a survey conducted by NTUC Learning Hub, SMEs are least likely to send their staff fo training as compared to larger private firms. Subsidised initiatives such as SkillsFuture can therefore help narrow this gap by encouraging SME leaders to build AI capabilities within their organisations and prepare their workforce for future technological change.
3. Government Grants as a Catalyst for SME AI Integration
Government grants have played an important role in helping SMEs explore AI while lowering the barriers to working with technology partners.
One example is a sustainability-focused non-profit organisation in Singapore that promotes Kampung culture. The organisation manages hundreds of volunteer shifts each week but requires a Volunteer Management System (VMS) to coordinate its operations more efficiently. Grants such as Singapore’s Productivity Solutions Grant (PSG) provide organisations like these with opportunities to work with technology providers and deploy digital platforms at subsidised costs.
While PSG primarily supports the demand side of technology adoption, other government initiatives such as Malaysia’s Digital Acceleration Grant (MDAG) focus on the supply side by supporting technology companies to commercialise AI solutions that can subsequently be deployed by SMEs.
Technology consultancies could leverage grant funding to deliver innovative AI projects to SMEs at subsidized early-adopter rates. These deployments provide insights that refine future grant rounds, ensuring programmes evolve based on real-world experience. This targeted approach directs support to sectors with the highest potential for AI-driven productivity gains, as reflected in the MDAG’s focus in nine priority sectors.
Conclusion
The costs of inaction are high. Whilst AI-Integration is nascent amongst enterprises, there is a significant risk of concentration amongst larger players who can afford the investment. As observed in the case of the OTA industry, early concentration of technology can favor a few players allowing them to set market standards in their own interest and marginalize SMEs. Policy makers must recognize this trend early and level the playing field with interventions that build ecosystems for SMEs to thrive. These interventions need to be agile in learning as they are deployed whilst respecting the sporadic nature of SME AI Maturity curve. Agile policy deployments can support AI innovation amongst SMEs resulting in healthy market competition, safeguarding employment opportunities and delivering quality to the end customers.
Acknowledgements
Case studies and observations in this article draw on industry observations and experiences in Indonesia and Singapore.
The views and recommendations expressed in this article published on August 2026 are solely of the author/s and do not necessarily reflect the views and position of the Tech for Good Institute.
