SUPERINTELLIGENCE SG .AIA first pass at the question: where should someone begin when they want to understand the country's AI ecosystem?
Singapore's AI landscape is not one list. Public programmes, research institutions, assurance work and adoption support answer different questions. This note puts those doors side by side so a reader can choose the next source to open.
The public record clusters around four jobs: build national capability, extend the research frontier, make systems testable and help organizations adopt AI in practice.
That structure is useful because it separates roles without pretending the boundaries are fixed. One institution can appear in more than one layer; this page records the public role described by its source.
Each row names a public source, the structure it reveals and the next question it leaves open.
AI Singapore describes a national programme that brings together Singapore-based research institutions and an ecosystem of start-ups and companies to perform use-inspired research, create tools and grow talent.
Read AI Singapore's public overview ↗Why it mattersThis is a useful entry point for capability, programmes and talent. It does not tell us every organization working in Singapore.
The Institute groups its work into foundational AI, AI + X, and AI governance and policy, linking technical research to sectors and public questions.
Read NUS AI Institute's research map ↗Why it mattersA research map helps a reader move from “who is doing AI?” to “what kind of question is being worked on?”
The Foundation describes an open-source community for AI testing frameworks, tools and practices, including the Global AI Assurance Sandbox for real-world generative AI applications.
Read AI Verify Foundation's public work ↗Why it mattersAssurance is infrastructure for adoption. It gives builders and deployers a place to ask how a system behaves, not only what it promises.
IMDA says the 2026 programme will support enterprises and workers, including a target of 10,000 enterprises and 100,000 AI-bilingual workers over three years.
Read IMDA's programme factsheet ↗Why it mattersAdoption is not only a vendor list. It includes skills, workflow change and the institutions that make implementation possible.
This is a public starting map, not a census. It does not yet show the full set of private vendors, community groups, funders, specialist labs or the lived outcomes of deployment. Those gaps are invitations for better sources, not missing rankings.