Business : YTL AI Labs, Nvidia Launch Open Dataset Reflecting Malaysia’s Population Diversity

YTL AI Labs has partnered with Nvidia to introduce an open-source dataset consisting of 1.35 million synthetic Malaysian personas. The dataset, called Nemotron-Personas-Malaysia, aims to provide developers and enterprises with a locally relevant foundation for building and testing AI systems. This dataset was created from 150,000 base records and reflects Malaysia's demographic, linguistic, cultural, and regional diversity.

It is freely available for commercial use under a Creative Commons Attribution 4.0 (CC-BY-4.0) license and does not contain any personal data. YTL AI Labs' CEO, Foong Chee Mun, emphasized the importance of representing Malaysia's unique languages, cultures, institutions, and working styles in AI systems. The collaboration between Nvidia and YTL AI Labs leverages Nvidia's synthetic data technology and YTL AI Labs' knowledge of Malaysian demographics and linguistic, cultural, and regional context.

The dataset can be utilized throughout the AI development process, from synthetic data generation to model fine-tuning, alignment, red-teaming, and application testing. This initiative aligns with YTL Group's sovereign AI strategy, which encompasses computing infrastructure, locally developed AI models, and data tailored to Malaysian users.