Economies from Japan to Vietnam stand to gain hundreds of billions of dollars in productivity through artificial intelligence (AI) by 2040. Open-source models may be key to ensuring smaller nations keep pace, according to a joint study by the Linux Foundation and Meta titled The Value of Open Source AI for APEC Economies.
The report examines 11 Asia Pacific Economic Cooperation (APEC) economies and reveals that 89% of organisations are already using open source in their AI strategies, primarily for its cost savings potential. Besides lower software ownership costs, they cite reduced vendor lock-in and improved productivity as key benefits.
That cost advantage is proving transformative for Asia Pacific, which is dominated by small and medium enterprises (SMEs) that lack the budget for expensive proprietary systems.
In Malaysia, where SMEs make up 97.4% of enterprises and contribute 38% of GDP, open source is seen as essential for competitiveness. “It levels the playing field and allows businesses without multimillion-dollar budgets to participate,” shares one Malaysian participant.
Even large institutions are adopting open source to sharpen their edge. Singapore’s DBS Bank deployed Red Hat OpenShift with confidential containers for its digital assets custodian operations. The move strengthened security while accelerating new service delivery. "Our collaboration with Red Hat has enabled us to innovate at greater speed and scale, providing secure and resilient services to our customers," says Ang Li Khim, group head of DBS Bank's Institutional Banking Group Technology.
Meanwhile, the Singapore Armed Forces' Digital and Intelligence Service is leveraging open-source cloud and AI for defence operations. "Cloud, software and AI have become crucial warfighting capabilities. We are deepening our partnership with industry innovators such as Red Hat to deploy and integrate leading commercial capabilities for operational edge," says ME7 Guo Jinghua, commander of the SAF C4 and Digitalisation Command.
See also: How AI agents can boost workforce performance and drive the bottom line
The shift underscores a growing consensus across the region that open-source AI can strengthen innovation.
"By opening access to advanced models and tools, open-source AI empowers businesses, educators, and public agencies to innovate faster and more responsibly. This collaborative approach enables APEC economies of all sizes to deepen their competitive advantage—helping advanced economies stay at the forefront of global AI leadership, while supporting emerging economies to leapfrog in digital development,” says Hilary Carter, Senior Vice President of Research at the Linux Foundation
The leapfrog effect
See also: Is Alibaba among the most undervalued AI companies?
Indonesia and Vietnam are testing whether open source can help emerging economies skip traditional growth stages entirely. Both countries have young, digitally native populations and are using accessible AI tools to build local solutions without massive capital spending.
According to the report, investment in Indonesian AI start-ups surged 141% between 2020 and 2024, and AI is expected to contribute US$366 billion to the economy by 2030. As for Viet Nam, it is projected to gain up to US$130 billion ($169.7 billion) in benefits from AI by 2040. This is likely supported by Viet Nam's National Innovation Center's partnership with Meta on Project ViGen to create open-source Vietnamese-language datasets for AI training.
"It's crucial that global corporations like Meta provide open source models that enable our businesses to develop and innovate," says Võ Xuân Hoài, the center's deputy director, as Vietnam targets becoming an AI nation by 2030.
Cultural customisation
Open source also enables localisation that proprietary systems often miss. For example, Singapore’s SEA-LION (Southeast Asian Languages In One Network) large language model lets developers build AI that reflects regional linguistic diversity and cultural norms. This is key for long-term adoption.
Building on SEA-LION, Indonesia’s GoTo and telecom provider Indosat developed Sahabat-AI, a repository of open-source models for Bahasa Indonesia and its regional languages. The project shows how open ecosystems can serve diverse communities.
Japan is taking a different approach. With nearly one-third of its population over 64, developers are adapting open models for elderly care, health monitoring, and robotic mobility. The country expects AI to unlock US$1.1 trillion in productive capacity.
To stay ahead of the latest tech trends, click here for DigitalEdge Section
Beyond social innovation, Japan is utilising open-source AI in scientific research and climate resilience efforts. The Japan Agency for Marine-Earth Science and Technology (JAMSTEC) fine-tuned Meta’s Llama model with climate-specific data from across the nation, from mountain villages to megacities. The result is an AI system that helps local governments make faster decisions in response to climate threats. By using open-source foundations, JAMSTEC built precise models without years of costly, ground-up development.
Infrastructure reality check
Although open source reduces software costs, the computational backbone still requires substantial capital. Respondents of the Linux Foundation-Meta study across Asia Pacific cited limited GPU access and compute capacity as major barriers, especially for smaller universities and rural communities.
Governments are responding with significant investments. Thailand has approved US$775 million for AI development through 2027. Malaysia set up a National AI Office in 2024. In the Philippines, the Department of Science and Technology plans to invest US$44.6 million in AI projects, in addition to the US$24 million already spent on R&D between 2018 and 2024.
The report also urges APEC economies to focus on workforce training and AI literacy. That includes upskilling and certification programmes to encourage responsible and secure use of open models, tailored to groups such as mid- and late-career professionals.
"Through this research, working with APEC regional experts, we've improved our understanding of how open-source AI can help address local challenges and unlock new opportunities for innovation. Working together, governments and industry can build open, trusted AI ecosystems by investing in local talent, digital infrastructure, and shared learning across APEC," says Simon Milner, Meta’s vice president of Public Policy for Asia Pacific.
