Singapore reviews high-risk AI safeguards as gambling regulator expands AI use

Singapore is assessing whether high-risk AI uses need tougher safeguards. In a written parliamentary reply on Tuesday, Minister for Digital Development and Information Josephine Teo said possible measures for essential services and other higher-risk applications include more rigorous testing, independently verifiable evidence, tighter deployment controls and stronger oversight. Teo said no cyberattack on government agencies involving an unsupervised AI agent has been reported, but the government does not dismiss catastrophic or extinction-level risks, nor does it assume they will materialize. It has not said when or whether additional safeguards will become mandatory.

Senior Minister of State Tan Kiat How said on Wednesday that frameworks for autonomous systems are under review. Guidance includes the voluntary Model AI Governance Framework for Agentic AI, introduced by IMDA in January and updated in May after feedback from more than 60 organizations. A Google-led AI Agents Sandbox with Singapore's Cyber Security Agency, GovTech and IMDA ran from August 2025 for about four months.

The Gambling Regulatory Authority's 2025/26 annual report called technology a strategic enabler. CEO Daniel Tan cited an AI Masterplan, expanded Microsoft Copilot use, AI bots for corporate resources and anomaly-flagging tools for audit reviews. The regulator renewed licenses for operators including Marina Bay Sands and Singapore Pools. Casino-related crime stayed low. Citizen and permanent resident casino visits fell to about 91,000, or 2.7% of the adult population, in 2025 from around 94,000, or 2.8%, in 2024, for the 12 months ended March 31, 2026. Levy holders were stable at about 6,700.