HÀ NỘI — As Việt Nam seeks new drivers of economic growth, artificial intelligence (AI) has emerged as one of the most promising technologies for boosting productivity, fostering innovation and transforming business models.

Yet AI is not a guaranteed engine of growth. Without adequate infrastructure, high-quality data, skilled workers and effective institutional frameworks, its adoption could instead worsen unemployment, widen inequality, increase security risks and deepen technological dependence.

The need to renew growth drivers and improve productivity has become increasingly urgent as Việt Nam pursues ambitious economic growth targets for the 2026–2030 period. At the same time, AI has expanded rapidly across virtually every sector. Unlike many technologies that directly affect only specific industries, AI has the potential to reshape almost every aspect of production, business operations and governance.

Hồ Đức Thắng, standing member of the National Assembly’s Committee on Culture and Education, told the online newspaper Thời báo Tài chính Việt Nam (Vietnam Financial Times) that the key to achieving double-digit economic growth lies in productivity, or creating greater value with the same resources — and no technology is better positioned than AI to drive such productivity gains.

At the macroeconomic level, former president of the National Economics University, Professor Trần Thọ Đạt and his research team developed three scenarios for AI’s contribution to Việt Nam’s GDP growth. 

Their model estimates that AI could add between 0.72 and 2.05 percentage points to annual GDP growth by 2030, depending on the pace of adoption and implementation. By 2045, the difference between the baseline and breakthrough scenarios could exceed two percentage points annually, highlighting the importance of policy choices and execution capacity.

The research also identifies four sectors — information and communications technology, finance and banking, science and technology, and manufacturing — that could generate more than half of AI’s total economic impact. These sectors already have relatively advanced digital capabilities, substantial data resources or strong potential to accelerate technology adoption across the wider economy.

If carried out effectively, AI could help Việt Nam transition from an input-driven growth model towards one increasingly driven by productivity gains and innovation.

The hidden costs behind productivity gains

Despite its enormous potential, AI adoption does not automatically translate into economic growth.

Director of the Institute of Vietnam and World Economy Nguyễn Đình Trúc said that the use of AI does not necessarily lead to higher productivity when operating costs are taken into account.

An individual might use AI tools to answer questions, draft documents or prepare presentations more conveniently, but such use does not necessarily translate into higher productivity at the company level, he noted.

Value is created only when AI is integrated into business and production processes, supported by reliable data, people capable of evaluating AI-generated outputs and clear accountability mechanisms, he added.

Without redesigning existing workflows, AI may simply become an additional cost or even introduce new errors if people place excessive trust in machine-generated results.

Thắng said the first major challenge lies in employment, noting that AI is not simply about replacing workers; it is reshaping how work is distributed.

Young employees entering the labour market may face the greatest pressure, as they typically perform entry-level tasks such as information gathering, report preparation, basic programming or administrative support.

Meanwhile, experienced professionals are often better positioned to leverage AI because they can distinguish accurate outputs from incorrect ones. As a result, AI may widen the gap between skilled and less-skilled workers, as well as between large firms with abundant data and resources and smaller businesses with limited access to technology.

Another bottleneck is the cost of AI infrastructure. Experts pointed out that developing large AI models requires enormous computing power and electricity consumption. These requirements are well beyond the investment capacity of most Vietnamese enterprises.

Experts warned that if Việt Nam relies entirely on foreign AI infrastructure, the economy could become dependent on external providers in terms of technology, costs and access to services. Conversely, attempting to build every foundational AI technology domestically could overstretch national resources and lead to fragmented investment.

Data serves both as AI’s raw material and as a source of risk. Historical data that is incomplete, unbalanced or biased may cause AI systems to reproduce or amplify existing inequalities. As more online content is generated by AI itself, using such material to train future models could gradually reduce the quality of AI outputs.

Additional concerns include AI-enabled fraud through deepfake images and audio, privacy violations, cyberattacks and the spread of misinformation. The pursuit of economic growth thus cannot be separated from the need for safety, transparency and accountability.

Moving from AI adoption to value creation

Early action will be essential, but this does not mean investing indiscriminately or following short-term trends. With limited resources, Việt Nam should focus on AI applications that can create measurable value, particularly in areas with strong economic impact and the potential to accelerate technology adoption across the wider economy.

At the national level, developing AI requires strong digital foundations, including computing infrastructure, reliable data systems and secure technology platforms. 

Đạt emphasised that the Government has an important role to play in investing in core digital infrastructure, particularly in areas related to network security, data safety and national digital sovereignty, while creating opportunities for private enterprises to participate in developing other digital technologies and solutions.

Building human capability is equally critical. AI training should go beyond teaching people how to use specific tools and focus on developing technical expertise, critical thinking, the ability to evaluate AI-generated outputs and a sense of responsibility in applying the technology. 

This requires both advanced training for highly skilled professionals and broader digital literacy programmes that enable workers and citizens to participate in the digital economy.

The opportunity is significant, but the benefits of AI will not be distributed automatically. Việt Nam needs to move early, invest strategically, and ensure that technology remains guided by human capability and responsible decision-making. — VNS