Beyond AI Literacy: Singapore Must Learn to Build with AI, Says BT
Source: Business Times
A Business Times opinion piece by Brian Lim and Leslie Teo argues that Singapore's focus on AI literacy is incomplete — the city-state must prioritise building AI products and companies, not just training workers to use the tools. While Singapore ranks first globally in innovation inputs, it only places ninth in outputs.
A Business Times opinion piece published Monday morning argues that Singapore's current focus on AI literacy programmes — teaching workers to prompt, use copilots, and fold AI into daily work — is necessary but fundamentally incomplete. Authors Brian Lim and Leslie Teo contend that the city-state must move aggressively from being a consumer of AI tools to a builder of AI products and companies.
The argument rests on a striking gap in Singapore's innovation performance. According to the World Intellectual Property Organization's Global Innovation Index 2025, Singapore ranks first in the world for innovation inputs — talent, institutions, and research. But it only ranks ninth for innovation outputs: patents filed, high-tech exports, and the global brand value of home-grown companies. As AI capabilities grow, the authors warn that the skill of operating a tool collapses in value. Knowing how to use AI, they argue, is becoming table stakes — like knowing how to use a search engine.
The piece highlights two Singapore-founded companies that exemplify the building approach. PatSnap assembled a proprietary corpus of over 190 million patents and trained its own domain-specific AI, compressing patent searches from weeks to minutes. Tookitaki uses federated learning to let banks in Asia pool anonymised money-laundering patterns without exposing customer data, improving detection models through collective intelligence. Both companies succeed not through generic AI capability but through proprietary data, trust, and domain-specific workflows.
Other governments are already moving past adoption. The Gulf is building at state scale. South Korea is racing on compute infrastructure. Vietnam is pulling in global partners while building local AI capability. In China, more than 20 cities now subsidise one-person AI companies, with Shanghai's Pudong district covering up to 300,000 yuan in computing costs for solo founders. The authors argue that Singapore's historical strategy of connecting flows and avoiding product risk, while sensible for a small open economy, risks hardening into a limiting mindset that prevents the city-state from capturing the full value of the AI era.
Why it matters for Singapore: The opinion piece lands as Singapore pours serious resources into enterprise AI adoption and workforce training through programmes like the S$1 billion NAIS 2.0 and Budget 2026's 100,000-worker AI training goal. But Lim and Teo's central challenge — that adoption alone won't create the companies, exports, and revenue that a mature AI economy needs — strikes at a deeper question about Singapore's economic strategy in an era where AI lowers the cost of building for the world from anywhere.