Business
Racing Ahead While Struggling to Monetize
Southeast Asia stands at a fascinating inflection point in the global AI revolution. While hyperscalers pour over $50 billion into regional infrastructure and adoption rates outpace global averages, a troubling paradox emerges from McKinsey’s latest research: companies are moving fast, but they’re not making money from it.
The Paradox
- Southeast Asia is rapidly adopting AI, with 46% of companies scaling implementations — above global averages.
- Despite heavy investment (over $50 billion from hyperscalers), 67% of firms report <5% EBIT impact.
- The issue isn’t technology, but execution and monetization.
The newly released “AI in Southeast Asia: An era of opportunity” report reveals a region sprinting toward an AI-powered future yet stumbling over the chasm between deployment and profitability. This disconnect should concern every C-suite executive, policymaker, and investor betting on Southeast Asia’s digital transformation.
The Adoption Illusion
The headlines look impressive. Nearly half (46%) of Southeast Asian companies have moved beyond AI pilots to scaling implementations, edging ahead of the global average and outperforming most of Asia-Pacific excluding China and India. With 680 million consumers, a population of 380 million under age 35, and mobile penetration reaching 930 million connections, the region appears primed for AI dominance.
Singapore alone hosts over 60 AI centers of excellence. AWS, Google, Alibaba Cloud, and Tencent have collectively committed tens of billions to data centers across Indonesia, Malaysia, Thailand, and Vietnam. The Southeast Asia-Japan Cable 2 went live in 2025, promising the low-latency connectivity that AI applications demand.
Yet beneath this glittering surface lies an uncomfortable truth: 67% of surveyed organizations report that AI has delivered less than 5% impact on their earnings before interest and taxes.
This is not a technology problem. It’s an execution crisis.
The Value Capture Gap
The McKinsey research identifies three structural barriers preventing Southeast Asian firms from monetizing their AI investments:
Barriers to Value Capture
- Talent shortage — lack of skilled AI professionals.
- Integration complexity — legacy IT and fragmented data hinder scalability.
- Unclear ROI — companies spend boldly but measure poorly.
First, the talent drought is real and worsening. Twenty percent of respondents cite lack of internal AI expertise as their primary obstacle, not budget constraints, not regulatory concerns, but the simple inability to find people who can make AI work. As Alexandro Seminiano, CTO at Bank of the Philippine Islands, notes: “We need people who understand the business and the context of the data being generated.”
Second, integration complexity is killing scalability. Sixteen percent of companies struggle to embed AI into existing systems, a problem compounded by legacy IT infrastructure and fragmented data environments that plague the region. AI isn’t plug-and-play; it requires fundamental workflow redesign that most organizations resist.
Third, the ROI remains unclear. Despite 64% of organizations allocating more than 11% of their technology budgets to AI initiatives, the business case for transformation remains murky. Companies are spending boldly but measuring poorly.
What High Performers Do Differently
The report’s most valuable insights come from studying the outliers, the 8% of Southeast Asian companies that have achieved full-scale AI deployment. These high performers share three distinguishing characteristics:
They redesign workflows fundamentally rather than layering AI onto existing processes. High performers are twice as likely to integrate AI at the core of operations, not the periphery. Grab exemplifies this approach: their merchant AI assistant, deployed to over 1.2 million merchants, has driven 10% business growth by embedding intelligence directly into seller workflows.
They invest boldly and consistently. High performers are 2.2 times more likely to expect enterprise-wide transformative change from AI, not incremental improvements. This isn’t about pilot projects; it’s about business model reinvention.
They embed rigorous AI governance. Nearly half of high-performing organizations demonstrate senior leadership ownership and commitment to AI initiatives, with formal governance structures that balance innovation with risk management.
The Agentic AI Wildcard
Perhaps most intriguing is the emergence of agentic AI, autonomous systems that act on behalf of users with minimal human intervention. Ninety percent of surveyed companies plan to experiment with AI agents in 2026, with IT (37%), software engineering (35%), and knowledge management (32%) leading adoption.
This represents a quantum leap beyond today’s generative AI applications. Yet scaling agentic systems beyond technical functions will require precisely the custom development and MLOps expertise that the region currently lacks. The companies that solve this capability gap first will dominate the next competitive cycle.
The Geopolitical Advantage and Risk
Southeast Asia enjoys a unique strategic position as the battleground where Chinese and American tech giants compete for influence. AWS’s $9 billion Singapore commitment, Google’s $2 billion Malaysian data center, Alibaba Cloud’s expansion across the region, and Tencent’s Jakarta operations create a competitive ecosystem that benefits local enterprises through choice, pricing pressure, and accelerated innovation.
As Mayank Wadhwa, President of Microsoft ASEAN, observes: “Southeast Asia is not just a consumer of AI; it’s become a massive co creator.”
Yet this geopolitical dividend comes with risks. International hyperscalers could inadvertently marginalize local innovation if governments fail to support domestic AI development. With over 1,200 languages spoken across the region, culturally-aware, locally-developed AI systems remain essential for inclusive growth.
The concerning reality: Southeast Asia’s AI start-ups received only $1.7 billion of the $20 billion in venture investment across Asia-Pacific in 2024, representing just 122 of 1,845 AI funding deals. While Q2 2025 saw venture investment jump to $172 million (the highest in three years), the capital gap remains dramatic compared to the scale of infrastructure investment by foreign tech giants.
The Micro, Small, and Medium Enterprise Challenge
The region’s economic backbone, MSMEs that contribute 44.8% to GDP and employ 85% of the workforce, face acute challenges in the AI transition. While platforms like Grab, Sea, and Shopee are democratizing access, smaller enterprises struggle with pricing pressures and capability gaps that threaten to create a two-tier economy of AI haves and have-nots.
Singapore’s minister for digital development and information, Josephine Teo, emphasizes the importance of leadership: “For AI to truly be transformative, leadership must drive the change. The CEO, C-suite, and board members all play a critical role.”
The Path Forward
The McKinsey report proposes a collaborative framework across five pillars: enabling trusted data flows, strengthening infrastructure and inclusion, expanding regional talent pipelines, catalyzing sector collaborations, and promoting responsible AI at scale.
These recommendations are sensible but insufficient without confronting hard truths:
Companies must stop confusing activity with progress. Piloting 50 AI projects doesn’t create value; scaling three transformative applications does. The discipline to kill experiments and double down on winners separates leaders from laggards.
Governments must balance openness with strategic autonomy. Attracting hyperscaler investment is necessary but not sufficient. Malaysia’s and Singapore’s investments in sovereign AI infrastructure represent the right instinct, retaining local capability while benefiting from global capital.
The talent crisis requires radical solutions. Traditional upskilling programs won’t close the gap fast enough. Singapore’s National AI Strategy 2.0 points the way, but the region needs aggressive immigration policies, stronger university-industry partnerships, and incentives for AI practitioners to relocate to Southeast Asia.
Value measurement must improve dramatically. If two-thirds of companies can’t quantify AI’s business impact, they’re measuring the wrong things. High performers obsess over outcome metrics, revenue growth, cost reduction, customer satisfaction, not deployment statistics.
A Region at the Crossroads
Southeast Asia’s AI moment is unfolding against a backdrop of genuine opportunity and legitimate concern. The fundamentals are strong: young populations comfortable with technology, competitive infrastructure investments, and healthy competition among global tech powers creating optionality for local enterprises.
But momentum without execution is just motion. The region’s 73% adoption rate means nothing if it doesn’t translate into the productivity gains, new business models, and inclusive growth that AI promises.
As Vivek Lath, McKinsey Partner, frames it: “Leading the AI charge in Southeast Asia requires bold, transformative ambition. It’s about moving beyond isolated use cases to fundamentally reinventing business models with AI at their core.”
The question isn’t whether Southeast Asia will adopt AI, the data shows it already is. The question is whether the region can close the gap between adoption and impact before competitors elsewhere figure out the formula first. With $4.12 trillion in GDP and 4.1% annual growth, Southeast Asia has too much at stake to settle for being fast followers who never capture the value they create.