SINGAPORE — On January 22, 2026, at the World Economic Forum in Davos, the Infocomm Media Development Authority (IMDA) launched the world’s first governance framework for autonomous AI agents. The framework establishes a structured approach to overseeing AI systems that can operate independently, make decisions, and interact with other agents without human intervention.

The new framework is part of Singapore’s broader AI governance infrastructure, which operates at three layers: framework, testing, and institutional. It builds on earlier initiatives, including the Model AI Governance Framework introduced in 2019 and a 2024 framework focused on generative AI. Singapore’s National AI Strategy 2.0 explicitly aims to position the country as a “trust anchor” in the global AI economy.

AI Verify, an open-source assurance toolkit governed by a non-profit foundation owned by IMDA, supports implementation of the framework. A company implementing AI Verify receives simultaneous credibility with Singaporean, American, and European regulators through a single audit pass. Singapore’s AI governance frameworks have been mapped to the U.S. AI Risk Management Framework, the European Union AI Act, and the international ISO 42001 standard.

Singapore’s AI Safety Institute leads the ASEAN Working Group on AI Governance, reflecting the country’s regional role in shaping policy. As of the time of reporting, Singapore had approximately 4,500 AI practitioners, with a government target to increase that number to 15,000 by 2029.

The 2019 Model AI Governance Framework provided principles for the ethical use of AI in commercial applications. The 2024 generative AI framework addressed transparency, safety, and accountability for large language models and other foundation models. The 2026 framework for autonomous agents extends this progression by focusing on systems that can act without real-time human oversight.