With an economy that has surpassed $500 billion in size, Vietnam is unlikely to achieve this target if it continues to rely solely on the growth drivers that have fueled its success for many years. To sustain double-digit growth, Vietnam needs a new engine capable of raising productivity across the economy.

The study by Prof. Dr. Tran Tho Dat and researchers from the National Economics University, presented at a national scientific conference on achieving Vietnam's development targets through 2030, has drawn considerable attention.

Rather than repeating familiar forecasts about AI, the research team focused on another question: How much can AI contribute to Vietnam's growth if the country pursues different policy choices?

Researchers currently do not have a unified voice on the impact of AI on growth. While McKinsey forecasts that AI can create enormous value for the global economy, Daron Acemoglu, the economist who won the Nobel Prize in 2024, argues that the impact of AI on productivity will be much more modest. 

Therefore, instead of looking for an absolute number, the research team chose to model the impact of AI under the specific conditions of Vietnam.

On that basis, the research team built three scenarios.

The baseline scenario assumes the economy absorbs AI according to a natural trend, with almost no major policy changes. The accelerated scenario focuses resources on industries with the highest AI applicability. And in the breakthrough scenario, AI is placed within a national strategy with synchronized steps on computing infrastructure, data, human resources and institutional regime.

The difference between the three scenarios does not lie in the AI technology itself. The technology will become increasingly common. The gap is determined by the country's level of preparedness to turn that technology into productivity and ultimately into growth.

That is also the big question that the study raises: in the new development stage, what will determine Vietnam's growth trajectory in the next two decades?

The answer lies in the gap between the three growth scenarios.

In the 2026-2027 period, the difference between scenarios is almost negligible. Investment in AI infrastructure, data, human resources or institutional reform all takes time to become effective. 

But after 2027, the growth trajectories begin to diverge clearly. The baseline scenario continues to grow according to the economy's natural AI absorption capacity, while the breakthrough scenario gradually creates distance thanks to a synchronously implemented national strategy.

The model's results show that by 2045, the difference between the baseline and breakthrough scenarios could reach about 2 percentage points of GDP. The cumulative gap between the two scenarios reaches about $340 billion, almost equivalent to the current size of the Vietnamese economy. That also means that delays in planning and implementing AI policies could cause Vietnam to trade off a very large part of long-term growth.

The research team therefore calls the period from now until 2028 the "golden moment". Each year of delay not only loses the growth opportunity of that year but also narrows the potential of the next two decades.

The model of Prof. Dr. Tran Tho Dat shows that what is measured is not only the impact of AI but also the impact of policy. With the same technology, the same starting point, different policy choices will create very different growth trajectories.

For that reason, the research team does not propose scattered investment. Among the 21 industries analyzed, four industries including information and communications technology, finance and banking, science and technology, and manufacturing are identified as priority areas.

Although accounting for only about 37 percent of GDP, these four industries are capable of generating about half of the economic impact of AI. When resources are not unlimited, these are the fields with the greatest spillover effect for the whole economy.

This approach is also in line with the spirit of Resolution 57, where science, technology, innovation and digital transformation are identified as important development drivers for the new phase. However, there is still a large gap from policy to growth. That gap can only be filled with specific policies on AI infrastructure, data, human resources, institutions and investment resources.

The research by Prof. Dr. Tran Tho Dat is not only about AI. What the research really points out is that with the same technology, the same starting point, different policy choices will create different growth trajectories.

Lan Anh