TL;DR: Vietnam has crossed 7 billion USD in announced AI data center investment, backed by deals from G42, FPT, Samsung, Viettel, and NVIDIA. A national AI Law took effect in March 2026, creating Southeast Asia's most comprehensive AI governance framework. Total compute capacity is on track to nearly double from 525 MW to 950 MW by 2030. For enterprises that rely on Vietnamese financial, company, and market data, the infrastructure shift matters now.
Why Is Vietnam Becoming a Regional Vietnam AI Data Center Hub?
Three structural factors are converging to make Vietnam AI data center investment one of the most active in Southeast Asia.
First, geography. Vietnam sits at the intersection of four major submarine cable systems linking Northeast Asia to Europe, the Middle East, and North America. The country's long coastline gives it low-latency connectivity that landlocked competitors cannot replicate. For any business running AI inference across Asia-Pacific, hosting compute in Vietnam cuts round-trip time meaningfully.
Second, cost. Vietnam's industrial electricity rate for large consumers runs substantially below comparable rates in Singapore, Japan, or South Korea. For energy-intensive GPU clusters, that gap translates directly into operating cost per inference call.
Third, policy momentum. The Vietnamese government set a national target of becoming a major digital economy by 2030 and has backed that target with legislation, funding mechanisms, and regulatory reform. The AI Law, national AI data center targets, and the National AI Development Fund are not aspirational statements; they are law with enforcement timelines.
The result: the Vietnam AI data center market was valued at 1.04 billion USD in 2025 and is projected to reach 3.18 billion USD by 2031, a compound annual growth rate of 20.5% (Mordor Intelligence, 2026). Physical capacity is expected to grow from 524.7 MW in 2025 to 950.4 MW by 2030.
The 7 Billion USD Vietnam AI Data Center Investment Map
More than 7 billion USD in Vietnam AI data center commitments have been announced as of mid-2026, spanning sovereign AI infrastructure, hyperscale facilities, and specialized AI factories (ERP.Today, 2026). Here is the project-by-project picture.
G42, FPT, and Viet Thai: 1 Billion USD Sovereign AI and Cloud
In February 2026, Abu Dhabi-based technology group G42 signed a framework cooperation agreement with FPT Corporation (Vietnam's largest IT services company) and Viet Thai Group to develop 1 billion USD in sovereign AI and cloud infrastructure across Vietnam. The deal covers three AI-ready data centers and a shared goal of building domestic AI capability that keeps sensitive Vietnamese data onshore (Bloomberg, February 9, 2026).
FPT's role in this Vietnam AI data center consortium is significant. FPT has operations in 30-plus countries and brings deep technical integration with the Vietnamese public sector. The sovereign framing, meaning compute and data assets controlled by Vietnamese entities rather than routed through foreign cloud providers, aligns directly with the government's data residency priorities under the AI Law.
NVIDIA and FPT: The 200 Million USD AI Factory in Hanoi
Separate from the G42 consortium, NVIDIA partnered with FPT on a 200 million USD AI factory in Hanoi. This facility has been operational since November 2024, making it Vietnam's most mature high-density GPU infrastructure. The NVIDIA-FPT AI factory provides GPU-accelerated compute for training and inference, and it is the primary reference point investors cite when evaluating whether the Vietnam AI data center ecosystem can support production workloads today, not just in 2028.
Samsung C&T and CMC: 1.3 Billion USD Hyperscale Hub in Ho Chi Minh City
Samsung C&T and CMC Corporation agreed a 1.3 billion USD hyperscale data center hub in Ho Chi Minh City, with a 30 MW first phase and a planned 100 MW expansion. CMC is already building a separate 250 million USD facility in the same city, with an initial 30 MW capacity expanding to 120 MW. Ho Chi Minh City's position as Vietnam's commercial capital and its proximity to undersea cable landing stations make it the natural anchor for Vietnam AI data center demand from financial services, e-commerce, and manufacturing sectors.
Viettel IDC: 200 MW Across Two Strategic Sites
Vietnam's national telecom operator Viettel IDC is developing two facilities that together exceed 200 MW of planned capacity. The Tan Phu Trung site in Ho Chi Minh City targets 140 MW, while An Khanh near Hanoi targets 60 MW. Viettel integrates its own AI solutions into both facilities, positioning them as full-stack AI data center environments rather than raw colocation space.
Other Major Vietnam AI Data Center Projects
The pipeline extends further. VDG is developing a 1 billion USD project; Kinh Bac City Development's Tan Phu Trung AI campus carries a 2 billion USD price tag; ST Telemedia Global Data Centres partnered with VNG Corporation (Ho Chi Minh City) on a 60 MW facility expected to complete in 2026. Together these projects account for the 7-plus billion USD figure cited across multiple independent sources.
Vietnam AI Law (March 2026): The Compliance Baseline That Changes the Game
Vietnam's National Assembly passed Law No. 134/2025/QH15 on December 10, 2025. It took effect on March 1, 2026, making Vietnam one of the first countries in Southeast Asia to establish a comprehensive, risk-based AI governance framework.
The law introduces a three-tier risk classification system. AI applications in finance, healthcare, justice, and labor are treated as high-risk. Providers of high-risk AI must meet pre-market conformity requirements before commercial deployment. For organizations already running AI tools in these sectors before March 2026, a 12-month compliance window runs until March 2027.
Article 20 of the AI Law provides that qualifying AI enterprises receive the highest level of incentives available under Vietnam's science and technology, high technology, digital transformation, and investment laws. This is not a minor rebate; it includes land use, import duties, and corporate income tax treatment.
The National AI Development Fund, to be established by the Ministry of Science and Technology in 2026-2027, will offer grants, loans, and preferential financing to domestic startups, SMEs, and foreign investors developing AI capabilities in Vietnam. An AI Voucher mechanism will support business adoption, and a regulatory sandbox will be available for AI solutions in sensitive sectors.
For enterprises building on Vietnam AI data center infrastructure, the AI Law creates a predictable operating environment. It also raises the compliance cost for offshore AI deployments that process Vietnamese personal data, creating a structural incentive to host AI workloads domestically rather than routing them through Singapore or Hong Kong.
What the Vietnam AI Data Center Boom Means for Enterprise Data Users
The infrastructure shift has direct implications for any organization that consumes or produces data in Vietnam. Here is how the Vietnam AI data center expansion translates into practical decisions.
GPU Compute Is Becoming Accessible to Vietnamese Enterprises
Until late 2024, accessing GPU clusters for AI training or high-volume inference in Vietnam meant either paying Singapore colocation rates or routing workloads overseas. The NVIDIA-FPT factory and the upcoming Viettel and CMC hyperscale facilities change that calculus. Vietnamese enterprises building AI models on Vietnamese company data, Vietnamese market data, or Vietnamese financial time series can now run those workloads domestically at competitive cost.
DataCore's Company Intelligence Service feeds into exactly this kind of workload: enterprise AI models that score counterparty risk, validate corporate identity, or enrich supply chain records rely on fresh, high-quality company data. Domestic compute capacity makes the training and inference cycle faster and more cost-effective.
eKYC and Identity Verification Enter a New Compliance Era
The Vietnam AI Law's high-risk classification for finance directly affects AI-powered Know Your Customer (KYC) and identity verification workflows. Organizations running AI-based eKYC pipelines must meet the new conformity standards by March 2027. The combination of domestic compute infrastructure and a clear regulatory framework means the compliance cost of running a compliant, onshore eKYC pipeline in Vietnam is decreasing even as the legal requirement to do so is increasing.
Knowledge Graph and Graph ML Workloads Benefit Directly
Graph-based AI workloads, including entity resolution, relationship mapping across corporate structures, and fraud pattern detection, are among the most GPU-intensive applications in enterprise data. The Vietnam AI data center buildout gives providers of graph intelligence services the ability to run these workloads on Vietnamese data at Vietnamese latency. For users of services like DataCore's Knowledge Graph Service and Graph ML Service, this translates into faster refresh cycles and lower query latency on Vietnam-specific entity graphs.
For more on how AI model infrastructure affects Vietnamese enterprise data services, see our analysis of US AI export controls and Vietnam's data infrastructure response. The regulatory and infrastructure dimensions are increasingly intertwined.
Vietnam AI Data Center Capacity Is on Track to Nearly Double by 2030
The scale of the Vietnam AI data center pipeline is not speculative. It is grounded in signed agreements and active construction.
As of early 2025, Vietnam had 41 active data centers with a combined capacity of 221 MW operated by 12 major providers including Viettel IDC, VNPT, FPT Telecom, CMC Telecom, NTT DATA, and ST Telemedia Global Data Centres. By end of 2025, the total had grown to approximately 525 MW across the expanded pipeline.
With Viettel's 200-plus MW buildout, CMC's 120-plus MW expansion, G42-FPT's three new sites, and the Samsung C&T hyperscale hub all progressing through 2025-2028, the Mordor Intelligence 2026 report projects total Vietnam AI data center capacity at 950.4 MW by 2030. That is near-doubling in five years from a already-growing base.
Deloitte's estimate for Vietnam's broader AI market reaches 65 billion USD by 2035, including 25 billion USD specifically from AI data center infrastructure. Whether that figure proves accurate depends heavily on execution; but the physical buildout now underway provides the foundation to support that trajectory.
The domestic cloud market context reinforces the trend. Vietnam's cloud market is projected to reach 1.24 billion USD in 2025 and 2.5 billion USD by 2029. Cloud and AI data center infrastructure are complementary: cloud capacity absorbs enterprise demand while AI data center capacity handles the GPU-intensive workloads that standard cloud instances cannot efficiently serve.
For a current view of the AI model landscape that DataCore's infrastructure supports, see our live AI model ranking for June 2026, which tracks 527 models across the DataCore leaderboard at airank.datacore.vn.
Vietnam vs Singapore: Is the Vietnam AI Data Center Landscape Competitive?
Singapore has been the default destination for Southeast Asian enterprise AI infrastructure for more than a decade. Its advantages are real: established regulatory certainty, mature power infrastructure, and deep connectivity. But several factors are shifting the competitive calculation in favor of a Vietnam AI data center strategy, particularly for workloads that process Vietnamese data or serve Vietnamese end users.
Land and power costs in Singapore have increased sharply. A data center moratorium from 2019 to 2022 constrained new supply, and even post-moratorium the government maintains strict sustainability requirements that raise operating costs. For a GPU cluster consuming 10 MW or more, the premium relative to Ho Chi Minh City is meaningful at the operating expenditure level.
Data residency requirements are also shifting. Vietnam's Personal Data Protection Decree (Decree 13/2023/ND-CP) and the new AI Law together create compliance conditions that increasingly favor processing Vietnamese personal data on Vietnamese soil. For a bank, insurer, or fintech operating under the State Bank of Vietnam's oversight, routing AI inference on customer data through Singapore introduces regulatory exposure that a Vietnam AI data center setup avoids entirely.
That said, Singapore's advantages in talent, legal infrastructure, and international connectivity remain real. The practical outcome for most large enterprises is a hybrid: Singapore for global or multi-market AI workloads, Vietnam AI data center capacity for Vietnam-specific use cases. This is consistent with the sovereign framing in the G42-FPT deal, which explicitly targets AI and data that should remain within Vietnamese jurisdiction.
How DataCore Uses the Vietnam AI Data Center Infrastructure
DataCore's data products and services run on Vietnamese data about Vietnamese organizations, people, locations, markets, and economic indicators. The relevance of the Vietnam AI data center buildup is direct.
DataCore's Knowledge Graph Service maps relationships across 500,000-plus Vietnamese corporate entities, using graph algorithms that are GPU-accelerated for both the training and the real-time query layers. The same applies to Graph ML Service, which detects anomalous patterns across corporate ownership and financial transaction networks. These workloads benefit from the lower-latency, lower-cost GPU compute environment that the Vietnam AI data center boom is creating.
DataCore's eKYC Service, which powers identity verification for financial institutions in Vietnam, is directly affected by the AI Law's high-risk classification for financial AI. The expanding Vietnam AI data center ecosystem means DataCore can maintain full data residency compliance while continuing to improve model accuracy on Vietnamese identity document patterns, a task that requires regular training updates that are both compute-intensive and legally sensitive.
For enterprise buyers evaluating DataCore's Company Intelligence Service, the practical implication is that the underlying infrastructure supporting real-time entity scoring and enrichment will become faster and more cost-efficient over the next three years as Vietnam AI data center capacity comes online. Latency and throughput improvements will flow through to API response times.
Frequently Asked Questions about Vietnam AI Data Centers
Is Vietnam's AI data center market open to foreign investors?
Yes. Vietnam actively encourages foreign investment in AI data center infrastructure. The 2026 AI Law provides the highest-tier investment incentives to qualifying AI enterprises, including foreign investors. The G42 (Abu Dhabi) and FPT deal is the clearest recent example: a sovereign wealth-linked Gulf technology group committing 1 billion USD in partnership with a Vietnamese corporation, backed by the government's AI strategy.
How does the Vietnam AI Law affect financial data services?
The Vietnam AI Law (Law No. 134/2025/QH15, effective March 1, 2026) classifies AI applications in finance as high-risk. Providers of AI-powered financial data or decisioning services must meet pre-market conformity requirements. Organizations already running AI tools in finance before March 2026 have a 12-month grace period ending March 2027 to reach full compliance.
Which cities are the main Vietnam AI data center hubs in 2026?
Ho Chi Minh City leads with the largest concentration of new projects. The Samsung C&T and CMC 1.3 billion USD hyperscale hub, the Viettel Tan Phu Trung 140 MW facility, and the G42-proposed 2 billion USD AI Factory are all anchored there. Hanoi hosts the NVIDIA-FPT 200 million USD AI factory (operational since November 2024) and Viettel's An Khanh 60 MW facility. Together the two cities account for the majority of Vietnam AI data center capacity.
When will Vietnam's total AI data center capacity double?
Based on current investment commitments, Vietnam's total data center power capacity is projected to grow from approximately 525 MW in 2025 to 950 MW by 2030, per Mordor Intelligence's 2026 market analysis. That represents near-doubling in five years at a compound annual growth rate of 12.6%. Several projects remain in early development stages, so the final 2030 figure could be higher if demand from AI workloads accelerates faster than the baseline forecast.
Sources
- ERP.Today: "Vietnam Draws More Than 7 Billion in AI Data Center Investments, Reshaping Enterprise Infrastructure" (2026). Link
- Bloomberg: "Abu Dhabi's G42 Leads 1 Billion Data Center Project in Vietnam," February 9, 2026. Link
- TechNode Global: "Abu Dhabi's G42 partners Vietnamese Consortium to form 1B AI data center in Vietnam," February 9, 2026. Link
- VietnamPlus: "Vietnam emerges as new hotspot for AI data centre investment." Link
- Vietnam Briefing: "Vietnam AI Law: Regulatory Milestone and Business Implications" (2026). Link
- Duane Morris Blog: "Vietnam: The First Law on Artificial Intelligence," March 3, 2026. Link
- Mordor Intelligence: "Vietnam Data Center Market Size and Share Outlook to 2031" (2026). Link
- GlobeNewswire: "Vietnam Data Center Market Investment Analysis Report 2026-2031," January 14, 2026. Link
Để lại một bình luận
You must be logged in to post a comment.