Skip to content

Mind the Gap: Can Southeast Asia’s Readiness Keep Pace with AI?

AI tools are spreading through Southeast Asia faster than governments can keep up, highlighted by a S$4.9 million deep‑fake scam in Singapore. High policy scores mask a lack of operational resilience and cross‑border coordination, prompting a call for dynamic governance and public education.

AI tools are spreading through Southeast Asia faster than governments and citizens can keep up. The risk became headline news when a deep‑fake scam in Singapore siphoned S$4.9 million after criminals pretended to be senior officials. That incident shows how high‑fidelity generative technology can move from a novelty to a financial threat in weeks.

Policy capacity versus operational resilience

Countries in the region have built strong policy frameworks, Thailand and Malaysia both scored above 80 out of 100 for policy capacity, but they fall short on resilience, with scores under 40. Singapore fared better on paper, ranking seventh globally in the 2025 Government AI Readiness Index, yet it still shares the same gap between rules and real‑world enforcement.

Indicator Figure Source
AI policy capacity (Thailand, Malaysia) > 80 / 100 fdef13c2‑869e‑47b6‑926e‑3ce98489edb6
Institutional resilience (same countries) < 40 / 100 fdef13c2‑869e‑47b6‑926e‑3ce98489edb6
Singapore AI readiness rank (2025) 7 / global 3ce134ba‑aff7‑4b56‑b7dd‑30b1c6cf2550
Deep‑fake loss in Singapore (May 2026) S$4.9 million 3933b3d6‑4a8a‑4e1d‑b966‑c86b637637a9
Gemini active users (Southeast Asia, 2026) doubled in one year dc202696‑7f61‑4b44‑b556‑e51040a742ac
Advanced Gemini achievement (2025) gold‑medal standard at IMO dfe26aeb‑19c0‑48fc‑8cce‑b9ee5bc7efe1
GPT‑4 maths accuracy (2023 Chinese test) ≥ 60 % across six grades 7f05d8c7‑99e9‑403c‑9996‑485a7c15e9fb

These numbers illustrate a common pattern: policy ambition is high, but the tools to apply those policies in real time are missing.

How exposed are people to AI‑driven scams?

A recent regional survey found that 65 % of digital‑native youth (aged 18‑34) and 44 % of older adults reported encountering a suspected scam. Confidence in spotting fraud is low, fewer than half of respondents felt they could reliably detect a scam, and an objective test measured an average identification accuracy of 71 %.

Demographic Reported scam encounters Confidence in detection
Youth (18‑34) 65 % < 50 % feel confident
Older adults 44 % < 50 % feel confident
Test accuracy (objective) , 71 %

The gap between exposure and ability to respond creates a fertile ground for AI‑generated deception.

Why governance must evolve now

Governments across Southeast Asia lack the operational machinery to turn high‑level AI policy into enforceable rules. They also miss the technical capacity to evaluate live models, translate guidance into actionable controls, and coordinate rapid responses across borders. The rapid uptake of AI, shown by Gemini’s user base more than doubling in a single year, amplifies the speed at which threats can spread.

Cross‑border coordination and system‑level safeguards are repeatedly cited as essential. Without a unified ASEAN approach, each jurisdiction is left to react in isolation, allowing scammers to exploit the weakest link.

Practical steps for organisations

Businesses can mitigate risk while public policy catches up. The following actions are a starting point:

  1. Audit AI‑related exposures, map where generative tools are used in your workflow and identify any external‑facing channels that could be spoofed.
  2. Strengthen verification procedures, require multi‑factor authentication and independent voice or video confirmation for high‑value transactions, especially when requests come from senior officials.
  3. Train staff on deep‑fake cues, run short workshops that demonstrate common manipulation signs, such as unnatural facial movements or mismatched audio‑to‑video timing.
  4. Monitor regional threat feeds, subscribe to alerts from Southeast Asian cyber‑security agencies that publish emerging AI‑driven scam patterns.
  5. Participate in ASEAN information‑sharing groups, engage with regional bodies that aim to harmonise response protocols and share forensic evidence.

These measures build a layer of resilience that policy alone cannot provide.

Closing thoughts

AI is reshaping the economic and social fabric of Southeast Asia at an unprecedented pace. The region scores well on paper for AI policy, yet operational resilience and cross‑border coordination lag behind. The deep‑fake loss in Singapore, coupled with widespread scam exposure, underscores the urgency of moving from static guidelines to dynamic, system‑level safeguards. By combining targeted public education, robust internal controls, and regional cooperation, Southeast Asia can narrow the readiness gap and harness AI’s benefits without falling prey to its darkest applications.

Terms used in this article

  • AI (Artificial Intelligence), computer systems that perform tasks normally requiring human intelligence.
  • Deep‑fake, synthetic media created with generative AI that imitates real people’s appearance or voice.
  • Resilience, the ability of an institution to absorb shocks and maintain core functions.
  • Policy capacity, the competence to develop comprehensive AI strategies.
  • Cross‑border coordination, cooperation between different countries to address shared threats.
  • Gemini, a generative AI product from Google mentioned in public reports.
  • GPT‑4, a large language model developed by OpenAI, cited for its performance on a Chinese maths test.