Vietnam Bets on AI to Navigate Next Transport Revolution Amid Hurdles

A decade after ride-hailing platforms reshaped Vietnam’s urban mobility, the country is poised for a new technological transformation driven by artificial intelligence. The previous wave of innovation altered travel habits, compelled traditional taxi companies to adapt, created new jobs, and forced regulatory updates.

“Today, a new technological wave, particularly AI, could trigger an even greater turning point for Vietnam’s transportation system,” said Nguyen Anh Duong, Head of the Department for General Economic Issues and Integration Studies at the Institute for Policy and Strategic Studies (IPSS). He was speaking at the “Mobility, Artificial Intelligence (AI) and Society: Strengthening the Future of Transportation in Vietnam” workshop.

Impacts on mobility

Artificial intelligence is already being integrated into Vietnam’s transportation and urban development policies. Modern cities are increasingly reliant on a new “digital infrastructure” comprising intelligent traffic operation centers, AI-enabled cameras, and data analytics platforms. According to recently published figures, Hanoi had installed over 1,800 AI cameras for traffic management by 2025 and intends to deploy more than 2,100 additional units this year.

The role of AI extends beyond surveillance to improving the detection and enforcement of traffic violations, thereby reducing the reliance on manual monitoring. Official reports indicate that in the first month of operation, Hanoi’s AI camera system recorded more than 6,300 traffic violations that resulted in administrative penalties. A similar system in Ho Chi Minh City identified over 3,400 violations in a single month.

Across Southeast Asia, nations are increasingly using AI to build cities more resilient to infrastructure pressures and environmental risks. Erlanggasakti Putra, a Program Analyst at the Tech for Good Institute (TFGI), noted that Singapore employs AI to predict congestion risks in its public transport system before they cascade across the network.

In Jakarta, Indonesia’s capital, AI assists with traffic navigation during extreme weather events like sudden flooding. The system can rapidly reroute vehicles to safer roads, mitigating public risk and maintaining mobility.

Simultaneously, the transition to electric mobility is amplifying AI's role in the region's green transport strategies. According to TFGI, Singapore aims for all newly-registered vehicles to run on clean energy by 2030 and plans a complete phase-out of internal combustion engine vehicles by 2040. Malaysia is targeting a major expansion of its electric vehicle (EV) infrastructure by 2030, with EVs projected to constitute 80 per cent of transport by 2050.

Indonesia aims to have 13 million electric motorcycles and 2.2 million electric cars on its roads by 2030 as part of its strategy to achieve net-zero emissions by 2060. Vietnam has also committed to net-zero emissions by 2050 at COP26 and has established a transition roadmap through Decision No. 876/QD-TTg, the Action Program on Green Energy Transition and Carbon and Methane Emissions Reduction in the transportation sector.

Experts believe digital technologies and AI will be critical to this process. “We often say AI is driving transportation development, but in reality, the growth of the transportation ecosystem is also feeding back and creating additional momentum for the AI sector,” Mr. Duong said.

Mr. Putra added that AI can support policymaking during the shift to electric mobility. Governments face complex, interconnected challenges, including electricity grid capacity, energy supply security, and the high upfront cost of EVs. He also stressed that AI's impact extends beyond technology to the labor market, where AI-based training and coordination tools can help workers in traditional transport sectors reskill for emerging jobs in the green economy.

Within the logistics industry, adopting technology at an appropriate scale can enhance operational efficiency and competitiveness. The Vietnam Logistics Report 2025, compiled by the Agency of Foreign Trade at the Ministry of Industry and Trade, notes that several leading firms, including the Sai Gon Newport Corporation, Viettel Post, and Vietnam Airlines, have integrated innovation and AI into their logistics operations.

Unlocking AI’s potential

The benefits of deploying AI in transport and logistics are becoming clear. Reports from Hanoi authorities show that AI-driven traffic management has markedly improved traffic flow and eased congestion. On some roads with AI systems, travel time on one-way routes has decreased by as much as 31 per cent, while traffic volume has grown by approximately 13 per cent.

Despite these gains, experts contend that several significant barriers must be overcome to realize AI’s full potential in mobility.

According to the IPSS, the first major obstacle is fragmented data. Information related to transportation is scattered across numerous agencies and organizations. Although Vietnam has enacted laws and government resolutions on data sharing and interoperability, connecting these disparate systems remains a formidable challenge.

“This is especially important for AI,” Mr. Duong explained. “AI development depends not only on algorithms or processing power but also heavily on access to data for training and operations. Fragmented and disconnected datasets have become one of the biggest barriers to AI deployment.”

A second challenge is the uneven distribution of AI capacity and infrastructure among localities. To date, most tangible progress has been concentrated in Hanoi and Ho Chi Minh City. Many smaller cities and medium-sized logistics firms face limited investment in AI and digital infrastructure, which risks widening the digital divide between regions and business groups.

The third challenge is cost. Building AI infrastructure for transportation demands substantial investment in camera systems, smart traffic signals, operation centers, data storage, processing infrastructure, and the energy to power it all. In addition to public investment, transport and logistics businesses face significant expenses related to their digital transition.

Experts warn that if technologies are too expensive, complex, or ill-suited to a company's scale, widespread adoption will be difficult, especially for small and medium-sized enterprises. While AI tools may seem affordable to individual users, the cost of deployment for businesses in transport and logistics presents a much greater hurdle.

On the policy front, Vietnam has established a relatively comprehensive and rapidly evolving AI framework. In 2021, the government issued Decision No. 127/QD-TTg on AI research, development, and application. More recent decisions on strategic technologies have further emphasized AI's role in national development.

In smart transportation, Vietnam has policies promoting the Internet of Things (IoT). However, experts point out that the emerging concept of the Artificial Intelligence of Things (AIoT) is largely absent from current regulations.

The key difference is in data processing. Traditional IoT systems collect data and send it to a central hub for analysis. AIoT, in contrast, allows for some data to be processed at the source via embedded AI, with only essential information sent to central systems. This approach speeds up processing, reduces latency, and boosts efficiency for smart transport networks.

Based on these realities, Mr. Duong proposed several priorities. He argued that Vietnam should first build an integrated and synchronized mobility data platform to support AI deployment. The country should also facilitate access to AI technologies for cities beyond Hanoi and Ho Chi Minh City to prevent a widening digital gap.

In parallel, Vietnam should accelerate the development of AI-based logistics corridors and expand regulatory sandbox mechanisms for new AI-powered services like autonomous delivery and automated operating systems. Without suitable sandboxes for both technology and policy, many new transportation models will struggle to advance beyond the pilot stage and achieve scale.