AI Safety & Ethics

Singapore treats “trustworthy AI” as a testable engineering problem, not a slogan. The flagship is AI Verify from IMDA and the AI Verify Foundation — a testing framework and toolkit built around 11 internationally recognised governance principles, with added evaluation and red-teaming methods for generative AI. The aim is to let companies actually test whether a model is safe and responsible, rather than simply assert it.

This hub gathers every parliamentary debate on AI safety and ethics, the related testing frameworks and guidance, and regulators’ statements — from model bias, hallucination and misuse risks to red-team evaluation, third-party audits and putting “responsible AI” into practice, so you can see how Singapore translates abstract ethical principles into operational tests.