Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
The Structural Challenge: Repeating Historical Patterns
Southeast Asia stands at a critical inflection point in its relationship with artificial intelligence technology. The region’s engagement with AI mirrors a centuries-old pattern: technological revolutions have historically concentrated wealth and control in the hands of early adopters and capital-rich economies, while peripheral regions assume subordinate roles in the value chain. Today’s AI revolution threatens to replicate this dynamic unless Southeast Asian governments and private sector actors deliberately reshape the terms of engagement.
The fundamental question confronting policymakers across ASEAN is not merely technical—it is fundamentally one of economic sovereignty and distributional justice. As AI systems increasingly mediate commerce, governance, and employment across the region, the decisions made in 2024-2026 will determine whether Southeast Asia becomes a genuine participant in AI development or remains confined to low-value implementation roles.
Ownership and Control: Who Builds the Infrastructure?
The current trajectory of AI adoption in Southeast Asia reveals a troubling concentration pattern. Major AI platforms and foundation models are overwhelmingly developed by technology firms headquartered in the United States and China. Companies such as OpenAI, Google DeepMind, and Alibaba control the algorithmic architectures that increasingly govern digital services across the region.
Southeast Asian governments have limited direct ownership stakes in the AI infrastructure upon which their economies increasingly depend. This contrasts sharply with the semiconductor manufacturing capabilities that Vietnam, Malaysia, and Thailand have developed over the past two decades. While these countries host significant portions of global chip production, they remain largely dependent on foreign intellectual property and design frameworks.
The absence of regionally-developed foundation models represents a strategic vulnerability. When Singapore’s financial regulators, Jakarta’s transport authorities, or Bangkok’s healthcare systems deploy AI solutions, they typically license technology developed elsewhere. This arrangement ensures that critical economic and governance decisions are mediated through foreign-controlled systems, creating both security dependencies and limiting the region’s capacity to shape AI development according to local values and priorities.
Labour Market Disruption and the Distribution of Work
Southeast Asia’s workforce faces acute exposure to AI-driven displacement. The region’s manufacturing sector, which employs millions across Indonesia, Vietnam, and Thailand, faces accelerating automation. Business process outsourcing—a sector employing over 1.5 million workers across the Philippines, India, and Vietnam—confronts disruption from large language models capable of handling customer service, data entry, and basic analytical tasks.
Unlike developed economies with robust social safety nets and workforce retraining capacity, Southeast Asian governments possess limited fiscal resources to manage large-scale labour market transitions. Indonesia’s unemployment rate stood at 5.2 percent in 2023; the Philippines at 5.5 percent. Mass displacement from AI automation could push these figures substantially higher, with particular consequences for rural workers and those without tertiary education.
The region’s demographic profile compounds this challenge. Indonesia, Philippines, and Vietnam have young, growing workforces that historically would have absorbed manufacturing employment. Instead, these cohorts may face a labour market where AI systems perform tasks that previously provided entry-level economic opportunity. Without deliberate policy intervention—including education reform, wage subsidies, or sectoral reorientation—AI could accelerate regional inequality rather than generate broad-based prosperity.
Value Extraction and the Surplus Question
The economic surplus generated by AI deployment in Southeast Asia flows overwhelmingly to foreign technology providers and the limited domestic firms that successfully integrate AI into their operations. When a Thai bank implements an AI-powered credit assessment system, the licensing fees and data flows benefit the technology vendor. When an Indonesian e-commerce platform uses AI recommendation algorithms, the algorithmic advantage accrues to the platform operator, not to the broader economy.
This surplus extraction dynamic resembles colonial-era resource extraction, albeit with different mechanisms. Rather than raw materials flowing to metropolitan centres, data and algorithmic value flow to technology companies concentrated in Silicon Valley, Beijing, and increasingly, Singapore. The difference is that data extraction occurs at scale and velocity previously impossible with physical commodities.
Malaysia and Vietnam have recognized this dynamic and begun investing in domestic AI capabilities. Malaysia’s AI roadmap prioritizes local talent development and sectoral applications; Vietnam has established AI research institutes and offered tax incentives for domestic AI startups. These efforts remain modest relative to the scale of foreign investment, but they represent recognition that passive technology adoption leaves the region economically subordinate.
Bearing the Costs: Social and Governance Risks
The costs of AI deployment—in terms of job losses, social disruption, and governance challenges—fall disproportionately on Southeast Asian populations. The region’s governance institutions often lack the capacity to regulate AI systems effectively. Data protection frameworks in most ASEAN countries remain underdeveloped compared to the European Union’s General Data Protection Regulation or emerging standards in Japan and South Korea.
Algorithmic bias in AI systems trained primarily on Western and Chinese data creates particular risks for Southeast Asian users. Facial recognition systems, loan approval algorithms, and hiring tools often perform poorly on non-Western populations. When these systems are deployed at scale—in banking, law enforcement, or public administration—the consequences fall on the populations least represented in the training data.
Misinformation amplified through AI-generated content poses acute risks in societies with polarized political environments. Indonesia and the Philippines have experienced significant social disruption from coordinated disinformation campaigns; AI-generated synthetic media could dramatically lower the barriers to creating and distributing false information at scale.
Strategic Pathways Forward
Southeast Asia requires a deliberate policy framework to reshape its relationship with AI from passive adoption to strategic participation. This framework should encompass three elements:
First, investment in regional AI capacity. ASEAN governments should establish joint AI research initiatives, similar to the European Union’s Horizon Europe programme. Singapore, with its advanced research infrastructure and financial resources, could serve as an anchor institution, but regional participation across Vietnam, Thailand, and Indonesia would ensure that local contexts shape research priorities.
Second, data governance frameworks that treat data as a strategic asset. Rather than allowing unrestricted data extraction by foreign companies, ASEAN should develop interoperable data governance standards that permit regional data flows while restricting extraction by non-regional actors. The European Union’s approach to data sovereignty provides a potential model, though adapted to regional circumstances.
Third, labour market and social protection policies aligned with AI deployment. This includes mandatory impact assessments for significant AI implementations, sectoral retraining programmes funded by technology companies, and progressive taxation on AI-generated corporate profits to fund social adjustment programmes.
Strategic Outlook: The Window for Autonomous Development
Southeast Asia has perhaps five to seven years to establish the institutional, regulatory, and technical foundations for autonomous AI development before the technology becomes so concentrated and embedded in regional systems that independent policy becomes infeasible. The region’s technological capacity exists—Vietnam produces world-class software engineers, Singapore hosts cutting-edge research institutions, and Indonesia’s data market represents enormous potential value.
The critical variable is political will. Establishing regional AI capacity requires sustained investment and coordination across ASEAN economies with competing interests. It requires resisting pressure from foreign technology companies seeking unrestricted market access. It requires acknowledging that passive technology adoption, while generating short-term efficiency gains, perpetuates long-term economic subordination.
The industrial revolutions of the nineteenth and twentieth centuries created lasting patterns of global inequality. Southeast Asia has the opportunity to ensure that the AI revolution produces different distributional outcomes—but only if policymakers treat AI governance as a core strategic priority rather than a technical issue delegated to regulators.
