CHUOR Porchourng
Researcher at the Cambodia Academy of Digital Technology
Question: Southeast Asia is a fast-growing emerging market with a distinctive pathway for AI development. What can Cambodia and other Southeast Asian countries bring to the network, and which areas of cooperation should be advanced first?
Cambodia and other Southeast Asian countries must prioritize foundational datasets that reflect local languages and cultures if they are to avoid marginalization in the AI era. Leading open and closed models perform relatively well in resource-rich languages such as English, Chinese and Korean, yet remain weak in Khmer and other low-resource languages and may reproduce cultural biases that do not fit local contexts. High-quality local-language data and transparent information about training resources remain scarce on public platforms. This affects not only model performance but also the ability of local societies to shape AI on their own terms. Closing the gap requires two capabilities to advance together: local talent able to curate and annotate data, train models and develop applications; and critical infrastructure, including GPUs and AI data centers. Data, people and compute are all indispensable. The capacity-building network should help member countries create these foundational assets so that AI addresses local needs and low-resource languages and cultures are not excluded from the global model ecosystem.
