Artificial intelligence is no longer a niche technology; it has become a structural force reshaping productivity, trade, public services, and labor markets worldwide. Governments that combine clear regulatory frameworks, substantial public investment, and private sector partnerships are better positioned to capture economic gains, while those that fail to coordinate risk falling behind.

This article compares how the European Union and ASEAN are approaching AI adoption, not as rivals, but as two separate policy ecosystems from which mutual lessons can be drawn. Europe has doubled down on a rule-based, infrastructure-heavy approach, characterized by sweeping regulations such as the EU AI Act (entered into force in August 2024, phased implementation through 2026–2027) and large-scale public investment like the AI Action Plan that mobilizes €200 billion to strengthen Europe’s AI competitiveness.

ASEAN, by contrast, is moving quickly on coordination, guidance, and practical capacity-building, as reflected in the ASEAN Responsible AI Roadmap 2025–2030 and the ASEAN Guide on AI Governance and Ethics 2024, both of which emphasize ethical AI deployment, capacity building, and regional interoperability rather than binding regulation. One concrete example of ASEAN’s digital-first approach is Malaysia’s Johor region, which has become one of Southeast Asia’s fastest-growing data center hubs, hosting major global investments in cloud and AI infrastructure.

The comparison that follows explores regulatory architectures, investment footprints, infrastructure, skills, and ethical drivers, before proposing lessons and collaborative opportunities that can accelerate inclusive AI-driven growth across both regions.

1. European Union

Regulatory Frameworks and Strategic Initiatives

The European Union’s cornerstone in AI governance is the EU Artificial Intelligence Act, which entered into force on 1 August 2024 and will become fully applicable in stages between 2025 and 2027. It introduces a risk-based system, banning certain applications such as social scoring, while imposing strict obligations on “high-risk” AI systems used in areas like healthcare, education, and law enforcement. Building on this legal framework, the European Commission announced the AI Action Plan in April 2025. The plan mobilizes an estimated €200 billion through the new Invest AI facility to strengthen Europe’s competitiveness in AI. These investments target sovereign AI model development, expansion of computing and cloud infrastructure, establishment of AI gigafactories to rival the U.S. and China, support for industrial and public sector AI applications, and large-scale digital skills programs to train the European workforce. Together, the AI Act and Action Plan illustrate Europe’s dual strategy: strict regulatory oversight to ensure trust and safety, combined with massive public investment to scale AI capabilities and industrial sovereignty.

Infrastructure and Investment

European governments and the European Commission are partnering with private companies to build AI fact‌ories and Centres of Excellence, combining research, industry development, and skills training, in order to support sovereign AI infrastructure and avoid overreliance on non-European providers. For example, the Euro HPC Joint Undertaking has selected sites in Finland, Germany, Italy, Luxembourg, Spain and other countries to host “AI Factories” that will offer AI-optimised supercomputers and experimental testbeds for model development and deployment inside Europe.

Another concrete case is the Barcelona Supercomputing Center (BSC-CNS), which received nearly €200 million to host one of the first AI Factories under the Euro HPC / EU Digital / Euro HPC funding schemes; the Factory includes expansion of the Mare Nostrum 5 supercomputer with AI accelerator hardware for training large-scale generative models for scientific and commercial use, and will serve startups, SMEs and public sector users for AI R&D and deployment. Private sector engagement is also showing up in collaborations such as NVIDIA’s recent announcement to work with European telecoms and industrial firms to deploy Blackwell-based AI infrastructure, build AI Factories in Germany to accelerate industrial and manufacturing applications, and establish AI Technology Centers across the continent to support research and upskilling of workforces.

Challenges and Criticisms

Even with strong policy momentum, Europe faces several gaps to close: significant disparities in AI adoption among member states, insufficient access to computing and data infrastructure, a shortage of digitally skilled talent, and fragmentation in regulatory enforcement. Critics argue that the EU AI Act’s prescriptive rules may slow startups’ innovation when compliance costs are high; for example, some early-stage AI firms estimate compliance obligations between €160,000 and €330,000 for basic regulatory requirements.  These costs can divert limited resources from R&D, slow time to market, and create barriers, especially for small firms lacking legal, technical, or financial capacity.

Which countries are doing relatively well? Italy and Spain lead in AI investment among EU members under the Next Generation EU Recovery and Resilience Facility, with Italy allocating approximately €1.895 billion and Spain approximately €1.2 billion to AI-related projects. On the other hand, Romania, Poland, and Bulgaria lag behind, with AI adoption rates among companies often below 5-6%, compared to Denmark, Sweden, and Belgium, which have rates of 25-28%. Another concern is regulatory fragmentation: although the AI Act aims for EU-wide harmonisation, enforcement and interpretation are expected to differ across countries, making cross-border scaling and compliance more complex for firms operating in multiple member states.

2. ASEAN

Policy Development and Regional Cooperation

The ASEAN Responsible AI Roadmap (2025–2030) was formally adopted on 5 March 2025, offering member states “customized and step-by-step guidance” to operationalize responsible AI in a coordinated and interoperable way across different national capacities. The roadmap is structured around four pillars: internal governance, skills & knowledge, risk mitigation & monitoring, and stakeholder coordination & regional cooperation. It also aims to set common principles (e.g. transparency, fairness, accountability) while leaving room for national flexibility. For example, the roadmap encourages member states to adopt AI risk impact assessments but does not prescribe a uniform threshold; instead, nations may calibrate assessment rules according to their economic, institutional, and technical readiness. Likewise, the ASEAN Guide on AI Governance and Ethics (2024), endorsed at the 4th ASEAN Digital Ministers Meeting, lays out baseline ethical norms while explicitly allowing countries to phase in rules or pilot governance models gradually. For example, Singapore’s Model AI Governance Framework has informed ASEAN’s guidance, while Vietnam, Thailand, and Indonesia are in various stages of building AI policy units and experimenting with generative AI oversight.

Innovation, Scalability, and National Actions

ASEAN countries are pursuing “cloud-first” and digital infrastructure strategies to undergird AI scaling. Malaysia has aggressively advanced its national cloud and AI infrastructure, notably through Johor’s data center investments.  In Indonesia, the government is prioritizing AI adoption in health and agriculture, with collaborations with private cloud providers to test AI use in rice yield forecasting. Thailand is likewise pushing AI in smart city pilot projects, tying them to its broader digital economy agenda. At the infrastructure edge, Malaysia’s Johor region has become one of Southeast Asia’s fastest-growing data center hubs, with total capacity (live, under construction, committed) exceeding 1,500 MW. The booming infrastructure has enabled hyperscale firms and AI model operators to host workloads locally instead of relying on distant regions, reducing latency and cross-border friction. Also, YTL Green Data Center Park in Kulai, Johor, is being developed in collaboration with NVIDIA to host AI infrastructure; YTL plans to power it with an on-site solar farm as part of its “green” data center ambition.

Challenges and Opportunities

ASEAN still faces persistent infrastructure gaps, as many member states lack consistent high-speed connectivity, reliable power, and cooling systems for dense AI workloads. The skills gap is severe: many countries struggle to supply qualified AI engineers and data scientists, especially in secondary cities and rural areas. Similarly, Data governance is also fragmented, and, as national approaches to privacy, data localization, cross-border flows, and cloud regulation differ widely, interoperability becomes complicated. Also, data sovereignty has become a pressing concern. Malaysia’s data center boom in Johor offers clear benefits for the local storage of sensitive information, yet it also poses potential risks of regulatory conflict should policy directions change, or resource limitations emerge. Additionally, environmental and resource pressures are mounting. Johor’s planned data center projects now represent roughly 5,800 MW of capacity demand, placing significant strain on local electricity grids and water systems. Moreover, the rapid pace of infrastructure expansion across the region highlights ASEAN’s broader challenge of aligning digital growth with sustainable energy development.

On the upside, ASEAN’s diversity is a comparative strength: countries can experiment with lightweight regulation in small states, offering lessons for larger ones; regional coordination allows for shared AI labs, cross-border sandboxes, and pooled infrastructure. Also, the projected AI market in ASEAN is expected to reach US$8.92 billion by 2025, rising to US$30.30 billion by 2030, showing the growth potential if these challenges are managed.

3. Comparative Analysis

Precaution versus Pragmatism

The EU represents a precautionary approach, embodied in the AI Act (2024), which classifies AI systems by risk and applies tiered obligations. Critics argue that stringent obligations for “high-risk AI” could slow down innovation and create compliance burdens for SMEs and startups. In contrast, ASEAN emphasizes pragmatism and flexibility through its 2025 Responsible AI Roadmap, encouraging member states to adopt AI governance principles in ways that match their institutional maturity. This divergence reflects differences in economic structures: the EU prioritizes consumer protection and rights harmonization, while ASEAN prioritizes developmental catch-up and innovation capacity.

Skill versus Speed

The EU is highly focused on skill development, such as the European Digital Education Action Plan, and coordinated AI research centres (e.g., ELLIS, CLAIRE) seek to build AI expertise at scale. ASEAN, while acknowledging skill deficits, prioritizes speed of adoption to integrate AI into sectors such as agriculture, health, and manufacturing. This difference arises because ASEAN economies are more sensitive to missed adoption opportunities, while the EU fears misuse of advanced AI without safeguards.

Culture and Ethical Considerations

Ethical foundations also differ. EU regulation foregrounds fundamental rights, such as privacy and non-discrimination, rooted in its constitutional and human rights frameworks. ASEAN emphasizes collective welfare, trust, and balanced innovation, reflecting its developmental priorities and cultural traditions of consensus-building.

4. Mutual Learning and Partnership

Regulatory Sandboxes and Pilot Programs

One promising area of mutual learning between Europe and ASEAN lies in the design of regulatory sandboxes. The EU’s AI Act establishes stringent obligations for “high-risk” AI systems, which may challenge smaller firms with limited compliance capacity. By contrast, ASEAN countries such as Singapore have pioneered AI sandboxes that allow firms to test algorithms under temporary, supervised conditions before being subject to full regulatory scrutiny. Expanding this model across both regions could serve multiple purposes: in the EU, it would reduce the burden of immediate compliance while still protecting citizens; in ASEAN, it would give regulators a structured way to monitor risks without freezing innovation. For example, a joint EU–ASEAN sandbox in the financial sector could allow fintech startups in ASEAN to trial AI-driven credit scoring tools under EU-style consumer protection rules, creating a learning channel for both sides.

The pilot program approach also lends itself well to sector-specific experimentation. Healthcare AI is an instructive case: cross-border pilots could focus on diagnostic AI tools for early cancer detection or infectious disease monitoring. The EU brings strong medical ethics frameworks and regulatory expertise, while ASEAN offers rich data contexts in tropical medicine and community health. Collaborative sandboxes would help refine ethical guidelines and technical safeguards that can then be scaled globally.

Joint Research and Investable Public Funds

A second recommendation is the creation of joint research funds that pool resources from both regions. The EU already operates Horizon Europe, a €95 billion R&D program, while ASEAN has begun mobilizing regional innovation funds. If these efforts were partially merged into a cross-regional AI innovation fund, it could target three urgent global priorities: climate resilience, healthcare, and cybersecurity.

For climate resilience, AI can be used to generate predictive models for extreme weather events such as typhoons or floods, which directly affect ASEAN, while also threatening Europe’s southern regions. A shared research fund would allow EU climate modeling expertise to integrate with ASEAN’s on-the-ground exposure and data, producing better tools for disaster preparedness. Similarly, in healthcare, ASEAN’s young and digitally active populations generate valuable health data, while Europe has advanced expertise in medical device regulation. A joint research program on AI-assisted diagnostics could improve global standards while lowering costs. Cybersecurity is another area where collaboration is essential: as ASEAN becomes a hub for global data centers, Malaysia’s Johor alone hosts more than 20 major projects, ensuring resilient and trustworthy AI-driven security systems are in the EU’s strategic interest, since European firms are among the largest investors.

Pooling funds would also attract private sector co-investment. For instance, an EU–ASEAN public fund could guarantee part of the risk in high-impact AI projects, encouraging venture capital to flow into areas that are socially beneficial but commercially uncertain. This hybrid funding model has precedent in the EU’s InvestEU program and could be adapted to the ASEAN context.

Skills Partnerships 

A third pillar of cooperation should be skills and capacity development. The EU has strong networks of universities and research institutes, such as ELLIS and CLAIRE, which are pushing the frontiers of AI science. ASEAN, by contrast, faces a critical skills gap, with only about 1 in 5 firms reporting access to sufficient digital talent. Bridging this gap requires innovative partnerships that go beyond traditional training.

One option is to establish joint fellowship and exchange programs where ASEAN engineers and policymakers spend time in European research labs, while European students and researchers gain hands-on experience with ASEAN’s high-volume, real-world data contexts, such as urban mobility in Jakarta or agriculture in Vietnam. Such exchanges would generate a two-way transfer: Europe gains exposure to scaling challenges in diverse environments, while ASEAN accelerates its technical expertise.

In addition, digital twin training programs could be co-developed. These would use simulated environments to train AI professionals on regulatory compliance, ethical dilemmas, and cross-border data governance. For instance, an EU–ASEAN “AI Governance Academy” could run joint courses online and in-person, drawing faculty from both regions and offering a credential recognized in both markets. This would not only build technical capacity but also harmonize professional standards, making collaboration smoother.

Joint Priorities with Mutual Value

Finally, the two regions should articulate joint priorities that yield clear and mutual value. One such priority is interoperable data governance standards. ASEAN’s data governance remains fragmented, with different sovereignty rules across countries, while the EU enforces strict GDPR rules. A harmonization initiative could focus on creating cross-border data trust frameworks, where ASEAN countries adopt lightweight standards compatible with EU norms, enabling smoother digital trade and investment flows.

Another priority is green AI infrastructure. As AI workloads grow, the energy costs of data centers become a concern. Co-investing in renewable-powered facilities in ASEAN, particularly in Malaysia and Indonesia, would benefit both regions: ASEAN would attract jobs and technological spillovers, while the EU would secure sustainable expansion routes for its tech firms. This aligns with both regions’ climate goals and can be positioned as a joint contribution to the global green transition.

Lastly, trustworthy AI certification should be advanced as a shared initiative. The EU’s conformity assessments under the AI Act could be piloted in ASEAN markets, adjusted for local capacity and economic structures. For ASEAN firms, this would open access to the EU’s single market by meeting compliance thresholds early, while for the EU, it would internationalize its AI standards. Such a framework could become a global benchmark, giving both regions disproportionate influence in setting norms for trustworthy AI worldwide.