According to Le Nhan Tam, CTO of Microsoft Vietnam, business leaders are exhibiting substantial interest in the advantages of AI.

Vetting and validation processes are vital to ensure AI accuracy and integrity, Source: Shutterstock

As evidenced by a Microsoft survey on AI work trends in which 200 business leaders were queried, 31 per cent of leaders prioritise “enhancing employee productivity” as opposed to “reducing the workforce” (16 per cent).

Additionally, Microsoft’s findings indicate that leaders harbour high expectations for AI, anticipating its ability to optimise underutilised human resources, elevate productivity levels, streamline operations, and facilitate more efficient work practices.

“In Accenture research, an overwhelming majority of survey respondents, up to 98 per cent, express a consensus on the significant role of AI utilising foundation models within their organisational strategies over the next 3-5 years. Furthermore, they hold the belief that the integration of AI applications will result in a substantial reduction of working hours by approximately 40 per cent,” Tam said at last week’s online workshop on the issue, held by the Vietnam Institutes of Directors (VIOD).

Tam believed that enterprises need to develop custom foundation models tailored to their manufacturing operations.

“For example, in the banking sector, it is imperative to refine the model based on the bank’s specific dataset. This process allows for the creation of an AI-like model that is highly attuned to the unique requirements and nuances of the banking business, thus maximising its suitability and effectiveness,” he noted.

However, it is also imperative for businesses to be attentive to potential risks associated with the implementation of AI in any technology system in order to minimise potential damages.

“Matters concerning reliable information, intellectual property, and information security are among the prominent concerns when discussing the utilisation of AI in particular and the broader application of IT in general,” he said.

Phung Viet Thang, Microsoft’s digital transformation consultant, emphasised that in the current trend of sustainable development, AI should be considered as an important part of the environmental, social, and governance (ESG) compliance assessment.

“However, according to a survey by the VIOD, more than 80 per cent of members of director boards (BoDs) are not ready for the AI control process,” Thang said.

How can AI assist BoDs?

Truong Manh Cuong, head of Strategic Business Development at SAP Vietnam, shed light on the exciting opportunities presented by new technologies and the incorporation of AI in business solutions.

Collaborative efforts with industry giants like Microsoft have allowed SAP to leverage big data and AI to cater to the operational and managerial needs of businesses.

“With more than 95 per cent of global automobile production currently managed through SAP’s systems, the prioritisation of integrating AI into business management processes remains a top concern for us,” said Cuong. “One notable product in SAP’s portfolio focuses on risk management within transactions involving stakeholders. By harnessing the power of AI, this solution enables intelligent procurement and supply chain management.”

He also explained how AI plays a crucial role in identifying and evaluating suitable suppliers based on internal transactional data and external connections.

“By analysing vast amounts of data and employing machine learning algorithms, AI can detect potential irregularities in transactions, providing decision-makers, including BoDs, with alerts regarding potential internal conflicts. This tool effectively facilitates fair processes and ensures informed decision-making,” Cuong said.

Furthermore, Nguyen Anh Tuan, chairman of Vietnam National Reinsurance Corporation (VINARE), also assessed how AI can greatly assist the BoDs in evaluating market conditions, monitoring competitors, and making rapid decisions and business development strategies amidst global fluctuations.

The utilisation of AI allows for timely access to updated information on input material prices, inflation rates, supply chain disruptions, and more.

“Instead of relying on quarterly board meetings based on pre-collected data, AI enables monthly meetings, ensuring that decisions are made in a timely manner,” Tuan said. “AI also proves valuable in generating management reports and bilingual English-Vietnamese annual reports swiftly, providing foreign investors with prompt insights into the company’s status.”

The ability to update these reports quarterly instead of limiting them to annual or biennial reports allows for greater transparency and comparability with enterprises from different countries, considering criteria such as ESG implementation, Tuan added.

The perils of inaccurate data

In the realm of corporate governance, ethical considerations reign supreme. For listed companies in Vietnam, the stakes are even higher as they face mounting pressures from regulatory bodies, shareholders, and the discerning eyes of the market.

Amidst this landscape, AI emerges as a powerful tool, but one that is not without its pitfalls.

Phan Le Thanh Long, CEO of VIOD, asserted that the successful implementation of AI hinges on the quality and integrity of the data input. AI thrives on data fed into it by human operators. However, the critical aspect lies in the freshness and accuracy of the information provided.

“Outdated or biased input data can generate flawed outcomes, undermining the very purpose for which AI is employed. The BoD should also consider the prevalence of chatbots that offer market-related advice, such as stock evaluations,” Long said.

These AI-driven systems rely heavily on data availability and accuracy to provide reliable insights. Yet, without ensuring the fairness, transparency, and ethical sourcing of this data, the resulting advice risks being skewed, potentially leading investors astray. Companies must recognise the vital importance of maintaining data integrity and ethical practices in order to harness the true potential of AI.

However, ethical considerations alone do not suffice in safeguarding the veracity of AI-driven systems.

Tam of Microsoft Vietnam emphasised the need for a meticulous approach to data input throughout the entire AI development process. It is imperative to foster an environment where AI grows and learns akin to human intelligence, but with a stronger emphasis on honesty.

Tam warned that erroneous data introduced at the inception of an AI system can have far-reaching consequences as the AI matures. Mistakenly informed decisions made by AI can be catastrophic, with potential detrimental effects on companies, public sentiment, and stock prices. The rise of deepfake technology further amplifies these concerns.

“Manipulated images, audio, and video can be used to fabricate false statements from corporate leaders, resulting in widespread misinformation and damaging the credibility of both the companies involved and the broader market ecosystem,” he noted.

To mitigate these risks, a multi-faceted approach is necessary. Rigorous vetting and validation procedures must be in place to ensure the accuracy and integrity of the data used to train AI models, according to Tam.

“Additionally, constant monitoring and evaluation of AI systems are vital, with a focus on identifying and rectifying biases or errors that may arise over time. Transparency and accountability should be core tenets guiding the development and deployment of AI technologies,” he said.

On the other hand, Tuan of VINARE also raised the matter of implementation of AI within the BoD and the responsible parties involved, highlighting the significance of legal regulations.

“According to the Law on Securities 2019, there are specific provisions pertaining to insiders within the organisational management structure. Additionally, the Law on Enterprises 2020 elucidates the scope of related persons,” Tuan said, adding that the relevance of the law to stakeholders, including the incorporation of AI, means that Vietnamese businesses are grappling with inquiries and seeking solutions to balance financial resources, investments, and human resources training.

“Questions arise about whether to implement AI internally or outsource, taking into account cost and schedule considerations. The control of related parties, partner authentication, competitor analysis, and information disclosure are also of significant importance,” he said.

“Furthermore, ensuring data and intellectual property security is crucial, as the implementation of AI presents challenges to the enterprise’s` business strategy.”