Hanoi (VNA) – Vietnam is entering a new phase of artificial intelligence (AI) development, shifting from adopting existing technologies to building its own capabilities. This requires simultaneous advances in models, computing infrastructure, data, and talent to create an ecosystem capable of developing technologies suited to Vietnam.

Strengthening core capabilities

At a recent meeting with Minister of Science and Technology Vu Hai Quan, Nvidia ASEAN Vice President and Managing Director for the ASEAN region Vikram Rao proposed four areas of future cooperation with Vietnam.

Nvidia will work with Vietnam to develop locally controlled large language models (LLMs) tailored to Vietnamese data, language and domestic needs, based on open-source models and Nvidia-developed tools. Nvidia has proposed that Vietnam develop a national plan for AI computing capacity to serve State agencies, businesses, universities, and research activities.

The company also proposed expanding its training programmes in Vietnam to a national scale and integrating them more deeply into curricula, while supporting scientists, postgraduate researchers and students in joining Nvidia’s AI ecosystem.

Another proposal is to connect Vietnam’s AI ecosystem with the global one, helping Vietnamese AI products and businesses gain deeper access to ASEAN markets and subsequently expand internationally.

These proposals align with Vietnam’s strategy for science, technology, innovation, and digital transformation. They embody a broader approach to AI development, moving beyond technology to individual products and services. To establish AI as a national technological capability, Vietnam must develop the capacity to research, evaluate, and operate models tailored to its data. This vision is articulated in the draft National AI Transformation Strategy to 2030, with a long-term outlook to 2045.

According to Dr. Tran Anh Tu, Deputy Director of the Department of Digital Technology and AI under the Ministry of Science and Technology, the strategy’s core principle is “comprehensive transformation through AI”. Rather than applying AI to isolated products, services or processes, the technology is expected to become a core capability spanning problem identification, system design and operations.

By 2030, the draft strategy targets AI support for all online public services and administrative records, with all ministries, sectors and localities adopting AI transformation plans. Some 60% of State agencies’ policy and management decisions are expected to be supported by data and AI.

In the private sector, AI is expected to support at least 500,000 small and medium-sized enterprises, household businesses, and cooperatives. Vietnam also aims to have at least 100 AI startups valued at more than USD 10 million. The draft proposes a national computing-capacity coordination platform, a Vietnamese-language model evaluation platform, at least five shared AI platforms, and eight Vietnamese-language models.

Overall, the strategy targets AI, contributing around 6% of GDP, and Vietnam ranks among the top three Southeast Asian countries in AI research and development. By 2045, Vietnam aims to rank among Asia’s top 10 countries in AI capabilities, with AI becoming a key operating platform for the State, economy and society.

This approach places AI within a broader capability-building framework that links models with data, computing infrastructure, and talent, and ultimately turns them into products, services, and businesses.

Prioritising data for AI

Lieutenant Colonel Dao Dinh Nam, Deputy Head of the General Advisory Division at the National Data Centre under the Ministry of Public Security, said AI can create sustainable value only when built on data that is accurate, complete, clean, live, consistent, shared and traceable.

The National Data Centre has identified priority data groups for AI development, including data for Vietnamese LLMs, computer vision, testing and evaluation, and safety and reliability assessment. These also include data for identifying fake news and misinformation, detecting AI-generated multimedia, testing attack scenarios and AI systems, and monitoring malware and cybersecurity threats.

AI-related physical infrastructure is growing quickly. Since the launch of National Data Centre No. 1 in August 2025 — the first national facility to earn top-tier international certifications for disaster resilience and security, big players like Viettel, VNPT, FPT, CMC, and VNG have been rolling out state-of-the-art data centres.

The developments reflect both growing competition in Vietnam’s data-centre market and rising demand from the digital economy. In 2025, Viettel announced two additional hyperscale data centres. There is one at Tan Phu Trung Industrial Park in Ho Chi Minh City, covering nearly four hectares with a designed capacity of 140 MW and around 10,000 racks, and another at the An Khanh Data Centre in Hanoi, covering 1.9 hectares with a total investment of 17.5 trillion VND.

Ho Chi Minh City is also attracting major data-centre projects, including the Evolution DC Vietnam HCMC project in the Tang Nhon Phu Ward High-Tech Area, with an investment of more than 508 million USD. It aims to develop large-scale data storage and processing infrastructure for cloud computing and gradually build exabyte-scale storage capacity. The city currently has 20 data centres and nine proposed projects, many targeting AI, cloud computing and growing data-processing needs.

Alongside models, computing infrastructure, and data, AI talent must be closely aligned with market needs. Recently, the Ministry of Science and Technology established the Global Vietnamese AI Expert Network to connect Vietnamese experts, scientists and entrepreneurs at home and abroad and mobilise high-quality talent.

Prime Minister Le Minh Hung has approved the National Programme for AI Human Resources Development to 2035, under Decision No. 1528/QD-TTg. The programme aims to develop AI talent at sufficient scale to support strategic technology sectors.

It targets the training of at least 50,000 university-level and higher-qualified personnel capable of applying AI in key sectors, as well as at least 10,000 highly qualified AI professionals, including researchers and engineers capable of developing advanced AI systems. At least 1,500 are to master core technologies and lead major AI projects.

To make businesses a central part of the AI ecosystem, the scheme on developing large-scale strategic domestic technology enterprises for 2026-2030, approved under Decision No. 1091/QD-TTg dated June 17, 2026, targets at least 10 major strategic technology companies with annual revenues of 1 billion USD.

Vietnam aims to rank among the top three Southeast Asian countries in AI research, development, and application, while developing at least six AI systems based on Vietnamese data for essential sectors including justice, finance, agriculture, the environment, culture, education, and healthcare. This will help move AI beyond experimentation towards commercially viable products and markets./.