Vietnam: AI Policy Brief — From Semiconductor Hub to Global AI Leader

Vietnam faces a unique AI policy opportunity: the country is simultaneously building semiconductor manufacturing capacity (Samsung $1.5B+, Intel $1.5B+, others $2B+) and establishing itself as the regional AI hub for Southeast Asia. With 101.6 million people, a median age of 31 (the region’s optimal workforce age), and a government that passed comprehensive AI legislation (AI Law No. 134/2025/QH15, effective March 1, 2026) in December 2025, Vietnam has positioned itself to compete with South Korea, Taiwan, and eventually China as a hardware-software-AI powerhouse. The strategic question is not whether AI will transform Vietnam, but whether policymakers can manage the transformation at the pace required without sacrificing the responsible innovation framework that justifies international investment.

Economic Exposure Assessment

Electronics & Semiconductors (6-8% of GDP, 2.4M direct jobs): Vietnam has become the world’s preferred alternative to Chinese semiconductor manufacturing. Samsung, Intel, TSMC (indirect), Qualcomm, and Apple (via contract manufacturers) have all expanded capacity. Vietnam’s share of global chip assembly and testing is currently ~1%, but the government targets 8-9% by 2030. AI is transforming this sector through quality inspection automation, predictive maintenance, and yield optimization. Unlike traditional automation that simply eliminates jobs, AI-enabled semiconductor manufacturing requires more skilled workers (technicians monitoring AI systems, engineers optimizing algorithms), but far fewer low-skill assembly workers. The sector will face intermediate disruption (2026-2027) but emerge stronger by 2028-2029.

Textiles & Apparel (3.5% of GDP, 2.8M jobs): Vietnam is the world’s third-largest garment exporter. AI is arriving through supply chain optimization (predicting demand, optimizing factory utilization), quality control automation (computer vision detecting defects faster than human inspectors), and design assistance (generative AI creating design variations from trend data). The apparel sector employs more workers than any other manufacturing sector, so even modest automation will affect hundreds of thousands of jobs. However, AI-driven supply chain optimization can increase responsiveness to market trends, potentially creating demand for higher-value goods that require more skilled labor.

Technology & Startups (2.8% of GDP, 280,000+ jobs): The only sector experiencing net job creation from AI. FPT Software, VNG, VinAI, and 278 AI startups are expanding employment faster than labor supply grows. The sector is attracting talent (and capital) at rates that suggest it could grow to 5-7% of GDP by 2030.

Financial Services (2.1% of GDP, 220,000 jobs): Payments, lending, and fraud detection are seeing rapid AI deployment. Viettel, VNG, and fintech startups are building AI-powered financial products. Traditional banking jobs (branch tellers, routine analysts) face disruption, but AI-literate financial roles (AI credit modeling, compliance specialists, customer intelligence analysts) are growing.

Agriculture (7.1% of GDP, 11.2M workers, 38% of workforce): AI adoption is minimal due to smartphone penetration limitations (72% in rural areas) and infrastructure gaps. However, this represents the largest opportunity: precision agriculture, AI-assisted crop selection, and climate-adapted farming could dramatically increase yields for the 6+ million smallholder farmers. The challenge is creating training and deployment infrastructure, not lack of technology.

Workforce Impact by Sector

Key insight: Vietnam’s unemployment rate of 2.22% means the labor market is tight. Job displacement will show up as wage pressure, not unemployment. Workers displaced from textile assembly or bank teller roles will shift to services, retail, or agriculture. The policy challenge is ensuring that displaced workers have pathways to retrain for higher-wage AI-related jobs before wage pressure becomes a macroeconomic constraint on growth.

Current Policy Assessment: The AI Law and Strategic Positioning

AI Law No. 134/2025/QH15: A Global Standard-Setter. Vietnam passed comprehensive AI legislation in December 2025, effective March 1, 2026. The law establishes a 4-tier risk classification: prohibited applications (social credit systems modeled on Chinese-style monitoring), high-risk (those affecting fundamental rights or safety), general-risk (consumer-facing AI), and minimal-risk (data processing tasks). This framework is more sophisticated and less restrictive than many proposed approaches globally. High-risk systems require impact assessments and transparency disclosures. Consumer-facing AI must disclose that users interact with AI. The law does NOT ban high-risk AI; it mandates responsible deployment. This balanced approach signals to international investors that Vietnam welcomes AI innovation within responsible constraints. The challenge is implementing the law without creating bureaucratic burdens that stifle startups.

Data infrastructure investment and NVIDIA partnership. Vietnam attracted $7+ billion in data center investment in 2025-2026 (Viettel $1B, Kinh Bac-AIC $2B, G42-Microsoft-FPT $2B+). NVIDIA’s decision to establish its Vietnam Research & Development Center as a third global AI hub (alongside California and China) signals that Vietnam is no longer a manufacturing follower but a technology leader. The VRDC will conduct research on AI systems and deploy infrastructure for Southeast Asian AI development. This is transformational but requires policy support: intellectual property protection, visa access for international researchers, and regulatory clarity on AI deployment.

Talent pipeline creation through FPT-NVIDIA partnership. FPT Software’s $200 million NVIDIA partnership aims to train 6,000 AI engineers by 2028. Combined with Hanoi University of Science and Technology’s specialized Generative AI program and government target of 8,000 AI graduates annually by 2030, Vietnam is creating AI talent supply at a pace that matches or exceeds demand projection. This is government-industry coordination done right: industry (FPT, NVIDIA) invests in training; government (Ministry of Education) aligns universities with curriculum requirements.

The gap: AI governance without stifling innovation. The primary policy challenge Vietnam faces is regulatory clarity without overregulation. The AI Law is well-designed, but enforcement mechanisms and clarification rules are still being established. Companies are operating with some uncertainty about what qualifies as high-risk AI, what data governance looks like in practice, and what transparency disclosures are required. Government should prioritize: clear guidance on the 4-tier classification, published examples of compliant and non-compliant AI systems, and a tiered enforcement approach that educates before penalties. Startups should know the rules; large companies should face consequences for violating them.

What Peer Countries Are Doing

South Korea: Established itself as a semiconductor and smartphone powerhouse, then pivoted to AI through companies like Samsung and LG. South Korea’s AI policy focuses on R&D investment ($2B+ annually) and talent education (Seoul National University, KAIST). South Korea’s advantage: mature semiconductor ecosystem. Vietnam’s advantage: lower labor costs, faster growth rates, and willingness to host foreign AI research centers (NVIDIA VRDC). Vietnam can potentially capture global AI investment faster than South Korea.

Singapore: Positioned as an AI governance and research hub through partnerships with tech giants (Google, Microsoft, Meta). Singapore’s advantage: advanced financial services, English-speaking workforce, stable regulatory environment. Singapore’s constraint: small population (5.7M) and high costs. Vietnam can serve as Singapore’s manufacturing and labor cost backup, while Singapore remains the regional governance and financial hub.

Thailand and Indonesia: Both larger than Vietnam in some metrics, but neither has committed to AI infrastructure investment at Vietnam’s scale. Thailand’s AI policy is nascent; Indonesia’s is fragmented across regions. Vietnam has seized first-mover advantage in regional AI infrastructure and governance.

Policy Recommendations

1. Establish an AI Implementation Agency (₫500 Billion / $15.6M over 3 years)

Create a dedicated agency responsible for: clarifying the AI Law’s 4-tier classification through published guidance and use cases; managing the transition for workers displaced by AI through retraining programs; coordinating between universities, industry, and government on AI talent pipeline; and monitoring compliance without excessive bureaucracy. This agency should have representation from FPT, VNG, semiconductor manufacturers, and civil society, ensuring that policy reflects both business and public interest.

2. Expand AI Education to Target 8,000 Graduates Annually by 2030

Partner with FPT Software, VNG, Viettel, and NVIDIA VRDC to develop curriculum frameworks for AI education at secondary, university, and vocational levels. Prioritize: computer science programs in all universities, vocational AI training for workers displaced from manufacturing, and specialized programs in hardware-software integration (given Vietnam’s semiconductor strength). Allocate 3% of national education budget to AI education, targeting 8,000 graduates annually by 2030.

3. Create an Agricultural AI Extension Service (₫1 Trillion / $31M over 5 years)

Agriculture employs 38% of Vietnam’s workforce but has adopted minimal AI. Deploy AI-equipped extension workers to 6+ million smallholder farmers with smartphone-based crop diagnostics, weather prediction, and market data. Partner with OneSoil, Cropwise, and other AI agricultural platforms. This could increase yields 15-25% while creating new jobs (AI specialists managing regional systems, training extension workers). Returns would exceed costs within 5 years through increased agricultural productivity.

4. Strengthen Data Protection and Governance (₫300 Billion / $9.4M)

The AI Law creates transparency and governance requirements, but data protection mechanisms are underdeveloped. Fund the Ministry of Public Security’s data protection agency with enforcement capacity: investigators, technical expertise, and penalties for data misuse. As AI systems collect biometric data, transaction histories, and health information on millions of Vietnamese, data protection must match the pace of AI deployment.

5. Establish Regional AI Research Centers in Hanoi, Ho Chi Minh City, and Da Nang (₫2 Trillion / $63M)

Vietnam can position itself as a regional AI research hub by establishing centers of excellence in: semiconductor-AI integration (Ho Chi Minh City, partnering with Samsung and Intel); fintech and financial AI (Ho Chi Minh City, partnering with VNG and Viettel); healthcare AI (Hanoi, expanding VinAI’s diagnostic platform); and agricultural AI (Da Nang or regional hub, serving Southeast Asia). These centers should attract international researchers through visa facilitation, housing support, and access to Vietnam’s data and hardware infrastructure. The goal: position Vietnam as the Southeast Asian destination for AI research, not just AI manufacturing.

6. Create an AI Startup Fund (₫3 Trillion / $94M)

Establish a government-backed venture fund focused on AI startups addressing local problems (agricultural AI, healthcare AI, supply chain optimization). Target deployment: ₫100-₫500M per startup, with focus on early-stage (seed to Series A). Success metrics: 20+ AI startups reaching Series B by 2028; 5+ reaching unicorn status by 2030. This transforms Vietnam from an AI employer market (people working for FPT, VNG) to an AI founder market (people building companies that might become global players).