Team Korea, consisting of researchers from Seoul National University, KAIST, and Naver, claimed third place in the Singapore AI Safety Red Teaming Challenge, hosted by Singapore’s Infocomm Media Development Authority (IMDA).

The challenge featured participants from nine Asian nations, including South Korea, China, Japan, India, Malaysia, Singapore, Indonesia, Thailand, and Vietnam. Competitors tested four large language models (LLMs) by crafting prompts aimed at exposing potential biases or discriminatory responses within the AI systems.