Vietnam needs to accelerate further in the race for Artificial Intelligence

The above comment was made by Mr. Nguyen Kim Anh, Chief Product Officer (CPO) of VinBigdata, when talking to VietNamNet about the Vietnamese language model that VinBigdata announced to have successfully developed on August 21. VietNamNet would like to send readers the content of this interview.

Mr. Nguyen Kim Anh, Chief Product Officer (CPO) VinBigdata.

Can you tell us about the current application of AI in Vietnam?

Mr. Nguyen Kim Anh: AI is one of the most popular keywords in the world in recent years. In Vietnam, we have also discussed this topic in many different forums. However, the reality shows that the implementation of AI in Vietnam still faces many challenges such as limitations in resources, infrastructure, professional human resources, etc.

The 2022 Artificial Intelligence (AI) Readiness Index report shows that Vietnam is ranked 55th globally, while Singapore is ranked 2nd, Malaysia 29th, and Thailand 31st. It can be seen that Vietnam still needs to be more determined in promoting AI research and application.

I think we have a lot of advantages in taking shortcuts because Vietnam has a young workforce with a very good foundation in Mathematics and Engineering. This is a prominent advantage that needs to be grasped. With the support and cooperation from the Government and businesses, I believe that we will quickly grasp the potential of developing the AI industry to contribute to the socio-economic development of the country.

Why did VinBigdata decide to develop the recently launched Vietnamese language model?

Mr. Nguyen Kim Anh: Large language models are a solid foundation for the creation of AI-based products, such as ChatGPT by Open AI or Bard by Google. However, for products from abroad, Vietnamese is not in the core language group. This affects the quality of content returned to users to some extent.

If you ask for detailed information, specific to Vietnamese people, the possibility of errors is quite high. More seriously, if users use this tool to look up information and update news, in the future it may lead to consequences such as incorrect basic knowledge of a part of Vietnamese people, especially young people.

Large language models are the solid foundation for generative AI applications.

Therefore, building a large Vietnamese language model is necessary to develop suitable products and best serve the needs of Vietnamese people. Right from the beginning, VinBigdata has identified "data" as the core and most important factor for researching and launching AI products to serve the market.

This is also the reason why we can pioneer the early completion of the construction of a large Vietnamese language model. Instead of having to use 175 billion parameters like ChatGPT, we only need a few billion parameters. This is not only meaningful in terms of application but also meaningful for society when the trend is that more and more people use generative AI as a tool for studying, working, looking up information...

With the newly announced Vietnamese language model, how will VinBigdata apply it, sir?

Mr. Nguyen Kim Anh: Mastering the technology completely, self-developing from the first steps, successfully building a large Vietnamese language model, is considered an important step to help VinBigdata bring generative AI technology into the ecosystem of products and services that have been and are being provided to the market.

We have outlined a specific roadmap for putting research results into practice. Specifically, VinBigdata will introduce two versions of the application to consumers, the first for businesses and the second for the community.

With the enterprise version, VinBigdata will integrate technology to make VinBase (a comprehensive multi-cognitive artificial intelligence platform) the first generative AI platform in Vietnam, while providing development solutions based on generative AI technology such as Generative AI chatbot, callbot or new generation ViVi virtual assistant...

This technology will help increase the naturalness of machine communication, while supporting users to search and synthesize information much faster and simpler than before.

Currently, VinBigdata has initially successfully applied new technology to the VinBase KB product line (VinBase Knowledge Base Portal). The product has the ability to retrieve information and automatically generate answers based on information collected from extremely large data sets in the knowledge system.

Successfully building a large Vietnamese language model is considered an important step forward to help VinBigdata bring generative AI technology into the ecosystem of products and services that have been and are being provided to the market.

According to our plan, by the end of December 2023, VinBigdata will launch the ViGPT product - "Vietnamese version of ChatGPT", open to the community for access and testing completely free of charge. With ViGPT, users can ask and answer questions about information specific to Vietnam (regulations, legal documents, etc.), or information of a specific nature (history, geography, literature, etc.).

This is a project of great significance to the community and society, expected to become a useful tool, effectively supporting Vietnamese people in the process of studying and working.

Does VinBigdata intend to share this large Vietnamese language model with interested parties?

Mr. Nguyen Kim Anh: The successful construction of a large Vietnamese language model is not only meaningful to VinBigdata but also meaningful to society and the science and technology community in Vietnam. The VinBase platform is integrated with generative AI, allowing businesses to fully utilize this core technology to apply to their own branded products.

With this model, VinBigdata hopes that Vietnamese businesses will have more opportunities to access AI technology to create or inherit achievements from the Vietnamese language model we developed.

In addition, with ViGPT, Vietnamese people will also have their own "Vietnamese version of ChatGPT", from which they can easily search for information sources quickly, accurately and reliably.

The Vietnamese language model was introduced by VinBigdata at an event in Ho Chi Minh City.

Security and copyright are important factors when developing generative AI. With the newly launched Vietnamese language model, how did VinBigdata handle this issue, sir?

Mr. Nguyen Kim Anh: As a research unit on big data and artificial intelligence, VinBigdata always respects the confidentiality of data through the development of strict processes and absolute compliance with data security regulations. Specifically, to ensure the rights of users, VinBigdata strives to create transparency in the process of collecting and using data, especially with personally owned data.

At the same time, we are always proactive in building multi-layered security measures, strict data lifecycle control processes, and appropriate data usage authorization (who has the right to access and use which data?) to prevent information leakage. There are also regulations and standards on data security in the world and in Vietnam, and VinBigdata always adheres to these principles.

In addition, copyright issues have also become a rather painful topic after the strong development of generative AI recently. Understanding this, VinBigdata focuses on developing products in the direction of providing specific, localized, and Vietnamese-specific information and in each specific field, rather than spreading it out to avoid misinformation.

Thanks to this initial orientation, VinBigdata can confidently create products that provide more accurate knowledge than ChatGPT in specific specialized fields such as finance and banking, law, medicine, etc. and with information that is indigenous to Vietnam such as history, geography, literature, etc.

In your opinion, is it time to apply AI into practice in Vietnam?

Mr. Nguyen Kim Anh: I think we need to accelerate further in the race for artificial intelligence. Because the world's major countries and corporations are already a certain distance ahead of us.

At present, it can be said that Vietnam has not really put AI into practice much and is only in the beginning stages.

However, I believe that with efforts and cooperation from all levels combined with high-quality AI human resources in Vietnam today, we can seize the opportunity to leapfrog!

Source