Singapore: Minister for Digital Development and Information released AI safety red teaming evaluation report

On 11 February 2025, the Minister for Digital Development and Information published the AI Safety Red Teaming Challenge Evaluation Report 2025 at the Global AI Action Summit in France. The evaluation assessed the performance of Large Language Models (LLMs) across different languages and cultural contexts in the Asia Pacific region. Conducted by the Infocomm Media Development Authority (IMDA) and Humane Intelligence, the challenge involved over 50 participants from nine countries testing four LLMs for cultural bias and stereotypes in non-English languages. The findings of this study aim to contribute to the development of consistent methodologies, benchmarks, and automated testing frameworks for addressing regional AI safety concerns.