A new research paper titled “AI Compute Sovereignty: Infrastructure Control Across Territories” was published on 20 June 2025 via SSRN, and is available to read for free (RM0).
The paper explores the growing importance of compute sovereignty, which refers to a country’s ability to independently manage, regulate, and control its AI computing infrastructure and supply chains.
This topic is particularly relevant to Malaysia as it strengthens its digital economy and expands AI capabilities through local data centers, cloud infrastructure, and public-private collaborations.
What is Compute Sovereignty and why does it matter?
According to the study, compute sovereignty goes beyond just having physical access to data centers or servers. It involves ensuring long-term control over the entire AI stack—including chip production, infrastructure, energy consumption, data protection, and policy enforcement.
The paper points out that many governments are increasingly aware of their dependence on foreign technology vendors—especially major hyperscalers like Amazon, Microsoft, and Google.
As a result, countries are now exploring ways to localise their AI infrastructure to protect their economies and national security interests.
In the Malaysian context, the issue of compute sovereignty raises questions about how much control local stakeholders have over infrastructure hosted by foreign companies, even when these facilities are physically located in Malaysia.
How this connects to Malaysia
Malaysia is actively positioning itself as a digital hub in Southeast Asia, with large investments from tech giants in building cloud and AI infrastructure. However, the SSRN paper suggests that hosting infrastructure alone is not sufficient for achieving true sovereignty.
To move toward compute sovereignty, Malaysia would need to:
- Invest in local chip design and production to reduce reliance on imported semiconductors.
- Develop regulatory frameworks that ensure national interests are preserved in AI development and deployment.
- Support local R&D efforts for compute infrastructure and energy-efficient AI systems.
- Build technical expertise in areas such as high-performance computing, cloud infrastructure, and AI safety.
- Secure access to critical resources, including clean energy and advanced manufacturing.
While Malaysia’s MyDIGITAL blueprint and recent partnerships with global tech firms help lay the foundation, the study highlights the importance of maintaining autonomy—not just participation—in AI infrastructure governance.
Core takeaways from the study
The study, authored by researchers from the Centre for the Governance of AI and the University of Oxford, outlines key observations that may shape Malaysia’s approach:
- AI requires immense computing resources that are currently concentrated among a few global players.
- Supply chain vulnerabilities—including access to chips and high-end GPUs—pose risks to national security and AI development.
- Countries must decide whether to build their own compute infrastructure or depend on foreign-controlled platforms.
- Compute sovereignty is becoming central to digital policy discussions globally, not just in the US or China.
These insights are timely as Malaysia begins discussions about local AI governance and long-term digital independence.
As Malaysia continues to digitise its economy, should it take more steps toward achieving compute sovereignty? Can the nation balance foreign investment with local control in its AI and cloud infrastructure? Stay tuned to TechNave.com for more updates.
COMMENTS