How can Vietnam participate more deeply in the global semiconductor value chain when it has fully prepared in terms of policies, legal framework, strategies, and science and technology programs?

Turning chip designs into finished semiconductor products still faces significant barriers.

According to the Ministry of Science and Technology, Vietnam's domestic semiconductor ecosystem currently includes about 60 chip design companies, 7,000 engineers and 166 universities involved in workforce training.

Despite great potential, research and tape-out infrastructure in Vietnam still has many limitations.

Research groups, universities, and companies mostly must place trial production at overseas fabs.

The cost for one transition from design to trial production ranges from $30,000 to $200,000, with waiting times of 12 to 24 months.

This reality makes research expensive, prolongs completion time, and limits product commercialization ability.

Therefore, identifying the right key products to avoid dispersing resources, while establishing shared infrastructure systems, is necessary to resolve this barrier.

Priority product groups

At the launch ceremony of the Vietnam National Multi-Project Wafer Coordination Center (VNMPW/CC), Vietnam's first national semiconductor prototyping hub, on June 26, leading semiconductor experts all agreed that Vietnam should not spread investment thin or choose too many directions at once.

Nguyen Thi Bich Yen, chairwoman of VSAP Lab, said, in the initial stage, focus should be on three groups with real demand, suitable to current capabilities, and with potential for future development.

“We don’t necessarily have to start with a head-on race at the most advanced nodes. Vietnam can start with simple functional chiplets, test vehicles, reusable IPs, package design, substrates, interposers, inter-die link testing, and post-packaging reliability evaluation,” she said.

The first group mentioned by Yen consists of Edge AI chips or neural processing units (NPUs) optimized for the Vietnamese language and specialized applications.

This group sees high demand across public administration, healthcare, education, transportation, and smart cities.

Instead of developing large-scale AI processors, Vietnam can start with small accelerator blocks that handle specific tasks like Vietnamese speech recognition or optical character recognition (OCR).

The second priority is microcontroller units (MCUs) and systems-on-chip (SoCs) designated for the Internet of Things (IoT) integrated with hardware security.

This product group capitalizes on matured technologies at reasonable costs, meeting data safety and device identification requirements for digital infrastructure.

The third priority is chiplets combined with advanced packaging.

Instead of designing a single large and highly complex chip, engineers divide the system into multiple smaller chiplets, each performing a specialized function, such as AI processing, memory management, security, data communication or power management.

Then, the chiplets do not necessarily have to use the same technology node; these chiplets are placed close together and reconnected in the same package thanks to advanced packaging technology.

Yen likened this technology to building a house. instead of constructing a massive building from the start, they build many good functional blocks, then connect them into a complex that operates synchronously.

Parts requiring high performance can use more advanced technology; parts only needing control, security, power management, or communication can use cheaper and more stable nodes.

No one can succeed alone

Beyond product strategy, experts stressed that transforming the way the semiconductor ecosystem operates is equally important. A strategic technology project cannot depend on a single research institute, university or company.

Minister of Science and Technology Vu Hai Quan said: "We cannot move forward alone or independently. We must establish clear partnerships.”

Building modern laboratory infrastructure, particularly cleanrooms, requires substantial investment that exceeds the financial capacity of many small and medium-sized enterprises.

To address this challenge, the government plans to establish shared testing centers and key laboratories across different economic regions.

The minister also raised three fundamental questions that will shape Vietnam's semiconductor ecosystem.

First, he highlighted the challenge of improving workforce quality as universities continue to face severe shortages of senior teaching experts, laboratory equipment and practical training facilities.

His second question focused on how Vietnam can move further up the global value chain. Policymakers need a clear understanding of the specific needs of small and medium-sized enterprises regarding manufacturing processes, materials and equipment so that the government can develop shared infrastructure that directly supports them.

Finally, the minister emphasized the need to identify Vietnam's flagship semiconductor products for the next stage of development, enabling the country to maximize existing technological capabilities while avoiding fragmented investment.

Thai Khang