Harnessing AI for Business: Practical Applications of AI in E-Commerce

Vietnamese businesses are racing to apply AI, turning this technology from a trend into a key factor in development strategy, despite many challenges.

Phuong Trang
This is some text inside of a div block.

30/9/2024 5:23 PM

As organizations in Vietnam and around the world are constantly striving to realize expectations about artificial intelligence (AI), this technology has gone from a “trend” to an important topic in high-level discussions of the Government through various forums.

According to the national strategy on research, development and application of artificial intelligence (AI) by 2030, Vietnam is aiming to join the group of four leading countries in the ASEAN region and the group of 50 leading countries in the world in the field of research, development and application of AI.

According to the Government AI Readiness Index report developed and published by Oxford Insights last year, Vietnam ranked 59th out of 193 countries and 9th in East Asia, with a score exceeding the regional average of 51.41.

This ranking assesses the readiness of countries around the world to exploit AI to effectively provide public services to citizens, including 39 indicators measuring digital capabilities, technological capabilities and data infrastructure. Compared to 2022, Vietnam has increased 19 places in the ranking.

However, domestic enterprises are also gradually realizing that deploying and applying AI is not an “overnight” task. Although in recent years, many Vietnamese enterprises have begun to incorporate AI into their management and operations to prepare for a future with AI playing a central role, especially in areas such as manufacturing, banking, retail, healthcare and e-commerce.

How to Sell Online

An "AI Readiness" survey published in November 2023 showed that only about 27% of organizations in Vietnam are seriously and comprehensively implementing AI.

The significant gap in numbers between organizations that have AI applications and those that are seriously and skillfully applying AI shows the challenges in implementing this technology.

Two prominent reasons to explain this are: businesses have difficulty balancing costs and developing effective AI technology exploitation strategies, along with macro challenges related to regulations and specialized human resources.

Challenges for businesses applying AI

In Vietnam, one of the biggest challenges in implementing AI is “unclear or lower-than-expected business performance”. Since last year, a series of businesses have been competing to apply AI for fear of missing out on this technology, but lacking a clear vision and plan.

To get on the right track, organizations need to clearly define the business challenge they need to solve or the business outcome they want to achieve through the application of AI as a superior technology, rather than just following a trend.

The most effective approach is sometimes not to have a next-generation GenAI model, but to bring AI into data analysis or prediction to speed up business processes and provide more reliable information to relevant employees.

Businesses need to understand that AI is not a goal to aim for. AI is essentially a technology that provides methods and tools to help businesses achieve their goals.

Another major challenge related to AI is the “high cost of ownership”. Since last year, the demand for graphics processing units (GPUs) to run AI has increased dramatically, leading to a global shortage of supply and increasing investment costs.

However, we don’t necessarily need GPUs to handle every AI application. Instead, there are many chips with different AI computing capabilities that are used for many AI use cases, such as central processing units (CPUs), large-scale integrated circuits (FPGAs), or application-specific integrated circuits (ASICs).

Harnessing AI for Business: Practical Applications of AI in E-Commerce

In fact, many common AI applications can be handled efficiently by CPUs without the need for additional GPUs. These CPUs can also be used for other tasks at the same time, such as networking and storage, which can save a lot of money and energy.

An equally important aspect is where the data is stored. According to Decree 53/2022/ND-CP issued by the Government, enterprises with branches and representative offices in Vietnam must store data locally on servers in Vietnam.

However, it is estimated that by 2025, 75% of the data generated will not be stored in data centers or on local clouds; but on edge devices (such as devices in factories or hospitals…).

This requires businesses to manage data storage in the most appropriate places: be it in the cloud, on-premises, or at the edge to handle the different requirements of inference and model training in the context of AI constantly evolving and expanding.

Additionally, as more AI applications move to the edge, there will be an increased need for AI PCs to leverage their improved processing capabilities for real-time data analytics and decision making.

Given this reality, businesses will need a flexible and scalable technology platform. A complex computing environment and open ecosystem that allows businesses to work with any vendor to choose the right hardware and software tools for their business needs will be the key to reducing costs while achieving high reliability, performance and better security.

Macro Challenges: Government Regulation and AI Expertise

Not only do businesses focus on promoting AI application and innovation, Vietnam also relies on the needs of the entire economy to promote AI development.

From a state management perspective, Vietnam has always been proactive in developing policies and encouraging investment in AI. The National Strategy on Research, Development and Application of Artificial Intelligence to 2030 allows Vietnam to promote research, development and application of AI in key areas such as security, defense, public services, resource and environmental management.

By 2025, the Government's strategy will call for the establishment of two national AI centers, ten research and training facilities, and AI will be "widely applied in public administration and online services to help reduce work processing time, human resources, waiting time and costs for people".

In addition to support from the Government, at the Vietnam Innovation Summit 2020 (VVS), 33 investment funds committed to pouring 815 million USD into innovative startup projects in Vietnam in the 2021-2025 period to develop the country's AI competitiveness.

Although the Vietnamese Government's AI readiness index is higher than the East Asia average of 51.41, like other countries, we are still in the early stages of exploiting the value of AI technology.

As governments around the world have yet to figure out the optimal regulations for AI, balancing protecting societal interests without stifling innovation is a challenging task.

Most importantly, we must adopt a flexible approach to AI based on fundamental ethical principles such as transparency, accountability, and fairness.

To make this happen, we can create an AI testing mechanism that allows organizations and businesses to experiment in a controlled environment before the technology is rolled out on a large scale.

From a socio-economic perspective, Vietnam’s large population and young generation give it an advantage in building a skilled workforce using AI. However, the gap between theory and practice still needs to be narrowed to help Vietnam truly achieve a competitive position.

According to the Government's orientation stipulated in the national strategy on research, development and application of artificial intelligence, to deploy basic skills on AI application and data science to contribute to promoting innovation for young people, Vietnam will focus on promoting the implementation of short-term and medium-term training courses on AI for students in different professions as well as types of human resources with career change needs.

In addition, the Government will also invest in building key research centers to train talents and high-quality human resources in AI and data science at leading universities and research institutes.

Another area to consider is expanding the understanding of AI among the non-technical workforce. With the rise of the low-code/no-code trend in developing AI tools and assistants, the ability of this workforce to grasp and utilize AI is also key to promoting the development of AI in Vietnam.

Becoming a leading AI nation

With many outstanding advantages, Vietnam is at a golden time to join the strong AI development trend in the coming years. By exploiting AI to improve product quality, optimize processes and enhance customer experience , Vietnam can create an unprecedented revolution in the way work is done. The continuous development of AI technology brings Vietnam a great opportunity to improve its overall competitiveness.

However, the AI ​​boom will also bring many challenges. As AI models continue to improve, AI approaches need to adapt and change. Therefore, the technology infrastructure, regulations in state and social management that are built and developed at the present time need to be flexible and scalable.

As the coming years will be an ideal time for countries to build AI platforms, Vietnam needs to improve public understanding of AI, ensure that AI is deployed safely without many obstacles from regulations, and build an open and diverse technology infrastructure. In this way, Vietnam can realize the great expectations from AI technology and make Vietnam a leading country in this field.