Southeast Asia's Digital Economy to Grow 17% to $368 Billion, Driven by AI
Southeast Asia's digital economy is forecast to grow 17% in 2026 to $368 billion in GMV, the total value of all goods and services sold through the region's online platforms. Generative AI is the main driver: 243 million people use it, and it is spreading four times faster than mobile internet did. 75% of respondents say they would let an agent compare prices and make a purchase for them. That makes a cancel button and clear spending limits essential for agentic shopping.
AI-processed from Google AI Blog; edited by Hamidun News
Southeast Asia's digital economy is set to grow 17% in 2026, reaching $368 billion in GMV. GMV stands for gross merchandise value, the total value of all goods and services people bought through online platforms: marketplaces, ride-hailing and delivery services, and travel sites. The figures come from e-Conomy SEA, the eleventh annual report, prepared by Google, Temasek and Bain & Company. According to the authors, the main reason for growth is generative AI: 243 million people in the region already use it, which is 58% of the working-age population. If you build products for Indonesia, Vietnam, Thailand or Singapore, these numbers help show where demand is heading.
What the authors counted
The report, titled "Building the Intelligent Economy," covers ten countries in the region. The authors estimate that generative AI is spreading four times faster than mobile internet did. By 2035, the region's digital economy is projected to exceed $1 trillion. Usage of AI, measured in tokens (units of text used to count the load on a model), is expected to grow 60-fold over five years.
By sector, the picture looks like this:
- e-commerce: $226 billion GMV, up 18% year over year; personal recommendations and in-app assistants lift conversion by more than 10%, and in some cases order volume rises 75%;
- online travel: $59 billion, up 10%; generative AI has overtaken social media and become the second most popular tool for planning complex trips, after standard search;
- online media: $44 billion in revenue, up 22%; engagement is 10–20% higher in quality terms, and advertising within commerce media grew by $4 billion;
- ride-hailing and food delivery: $40 billion, up 15%;
- digital financial services: $1.65 trillion in transaction volume, up 16%; AI helps detect fraud in real time and expands access to credit.
People are ready for agents if the purchase can be reversed
The most useful data for developers and product managers concerns trust. 72% of users already rely on automated features such as price tracking. 75% agree that an AI agent may compare prices and pay for a purchase on their behalf. 62% trust AI tools more than they did a year ago, citing proven experience (56%), habit (55%) and accuracy of answers (53%).
But the consent is not unconditional. To feel comfortable, people want privacy protection (44%), a clear explanation of what the AI can and cannot do (36%), and the ability to cancel a deal made by an agent (30%). This amounts to a ready-made checklist: agentic purchases need a refund button, clear spending boundaries ("do not spend more than this amount"), and an activity log the user can review.
What holds growth back
The authors name four conditions. First, businesses need to move from answering single questions to handling multi-step tasks end to end, and model deployment must become cheaper so that small companies can afford it too. Second, a 60-fold rise in consumption will require roughly three times more data center capacity over five years, with AI accounting for most of it.
Third, energy efficiency. Singapore is introducing mandatory requirements for PUE (the ratio of a data center's total energy to the energy consumed by the servers themselves) under its digital infrastructure law, while Malaysia is developing the CRESS scheme for purchasing clean energy. Fourth, unified trade rules.
The ASEAN DEFA agreement is expected in November 2026, which the authors describe as the world's first regional digital trade pact.
How to read these numbers
The report should not be treated as independent: Google has a stake in the growth of the region's AI market, and forecasts to 2035 remain forecasts. Moreover, the 75% who are ready to delegate purchases to agents are survey answers, not actual payments, and the gap between "ready" and "did it" is usually large in e-commerce.
The direction is nonetheless clear. The market is large and fast-growing, consumers are broadly receptive to AI, and the bottlenecks lie in infrastructure and energy. Teams entering the region with agentic shopping, support or financial services would be wise to build in operation reversal and transparency from day one. For those selling compute and cloud services, the report signals demand: roughly three times more data centers over five years.
- Nov 21, 2026Apple Is Building an AI Agent and Making It Harder for Outsiders to Get Access: Permission Now Takes 14 Steps, Not 4
- Nov 21, 2026Gemini 4 Neon shrank the TS Rust binary from 47.44 to 7.97 MB in under 10 hours
- Nov 21, 2026A swimming team captain handed ChatGPT three scheduled tasks and cut her routine work by a multiple
- Nov 21, 2026MIRA Neural Network Renders Rocket League at 20 FPS, but Ping Hits 460 ms
Key Questions Answered
What the authors counted?
The report, titled "Building the Intelligent Economy," covers ten countries in the region. The authors estimate that generative AI is spreading four times faster than mobile internet did. By 2035, the region's digital economy is projected to exceed $1 trillion. Usage of AI, measured in tokens (units of text used to count the load on a model), is expected to grow 60-fold over five years.
What are the key details on People are ready for agents if the purchase can be reversed?
The most useful data for developers and product managers concerns trust. 72% of users already rely on automated features such as price tracking. 75% agree that an AI agent may compare prices and pay for a purchase on their behalf. 62% trust AI tools more than they did a year ago, citing proven experience (56%), habit (55%) and accuracy of answers (53%). But the consent is not unconditional.
What holds growth back?
The authors name four conditions. First, businesses need to move from answering single questions to handling multi-step tasks end to end, and model deployment must become cheaper so that small companies can afford it too. Second, a 60-fold rise in consumption will require roughly three times more data center capacity over five years, with AI accounting for most of it. Third, energy efficiency.
Ship AI Faster with Founders, Engineers & Weekly Architecture Reviews
Inside Vibe Club (hamidun.com/club) — weekly live masterminds with Zhemal Khamidun, Tech Radar, full LMS access, and 370+ production Claude Code skills.
The AI world, distilled — once a week
Seven stories that actually mattered, hand-picked. No noise, no reposts, no press releases.
Done! Check your inbox for a confirmation.