Southeast Asia Pursues AI Autonomy Amid US-China Tech Rivalry
Southeast Asian nations are adopting Cold War-era hedging strategies to develop independent AI capabilities. This aims to reduce reliance on US and Chinese technology stacks, despite significant challenges.
Hedging Against Tech Dominance
Southeast Asian countries extend Cold War-era political hedging into the AI development race. They seek a third pathway, independent of both the United States and China.
The US promotes its full AI technology stack globally, including hardware, software, data systems, and models, as a "gold standard." An October paper by John Lee, published by ISEAS Perspective, argued for developing a niche outside US-China integration of AI and robotics.
China, meanwhile, invests heavily in domestic AI capabilities, aiming to replace the US as the dominant actor and shape regional AI standards.
National Initiatives for "Third Stack"
Responding to this competition, Southeast Asian nations explore a "third stack" movement. This involves domestic AI stack development and new policy frameworks to establish sovereignty over AI data streams. Vietnam introduced the region's first AI law and a Law on the Digital Technology Industry.
Hanoi classifies digital systems as strategic assets, subject to localised data requirements. State-linked FPT uses US Nvidia chips for core compute but deploys models via its sovereign FPT Smart Cloud. Private firms like VinAI and Semikong developed localised Large Language Models (LLMs), such as PhoGPT, with FPT software, reducing foreign LLM reliance.
Singapore and Regional Efforts
Singapore also uses public policy to achieve greater domestic control over AI stacks. In January, it launched the Model AI Governance Framework for Agentic AI, updating its 2024 Generative AI policy. State-linked AI Singapore developed the SEA-LION family, an open, multimodal AI model.
This addresses the under-representation of Southeast Asian languages and cultures in frontier AI models. Malaysia, Indonesia, and Thailand also seek greater control over domestic AI systems by balancing US and Chinese AI investments and developments.
These national efforts show progress toward technological autonomy, but significant hurdles remain. Southeast Asian nations still rely on American or Chinese software and hardware. Jacob Taylor and Joshua Tan note that national efforts cannot match the scale of compute, talent, or high-quality data of US or Chinese AI ecosystems.
Developing fully independent AI stacks is economically difficult. Current efforts often result in localisation of existing foreign AI models through technology transfers. Instead of purely national strategies, collaborative regional hedging, perhaps modelled on Airbus, offers a more scalable path for developing AI products under a public utility model.
This article is journalism, not investment advice; consult a licensed professional before making financial decisions. Market data is indicative, may be delayed, and should be verified with your broker or exchange before use.
Comments.
Comments are moderated. We remove what is unlawful, abusive or off-topic, and and you remain responsible for what you post.
Reader comments open soon. Until then, corrections and responses go to our newsroom, and we publish what we get wrong on Corrections.