含羞草社区

Singapore is accelerating its push to become a trusted hub for artificial intelligence (AI), with the Singapore Budget 2026 allocating significant resources towards AI development, infrastructure and enterprise adoption.

Businesses across sectors are increasingly integrating AI tools into customer-facing services, internal workflows and decision-making processes. This has increased business efficiency and resulted in more competitive businesses.

Singapore AI risks and liability

That said, the risks associated with AI are not theoretical. In Singapore, two lawyers were recently sanctioned for filing court submissions that cited non-existent cases generated by AI tools.

This underscores a broader concern: the reliability of AI outputs and the consequences of over-reliance without adequate human oversight.

Similar risks have also arisen in other commercial contexts. In one widely reported case, an airline was held liable for inaccurate information generated by its chatbot and provided to a customer, highlighting that businesses may ultimately bear responsibility for AI-driven interactions.

The Singapore Regulatory Approach

Unlike jurisdictions such as the EU, which has introduced a comprehensive, risk-tiered regime under the Artificial Intelligence Act, Singapore has thus far adopted a more calibrated, sectoral approach.

Regulatory guidance has been issued through initiatives such as the Model AI Governance Framework, developed by Singapore’s Infocomm Media Development Authority, alongside industry-specific guidelines. Whether Singapore will eventually enact overarching AI legislation remains to be seen.

That said, the absence of a single, omnibus AI statute does not mean that businesses are free to deploy AI without constraint. Existing legal frameworks continue to apply with full force.

Depending on the use case, AI deployment may give rise to liability under principles of misrepresentation, negligence or breach of contract. Where personal data is involved, obligations under the Personal Data Protection Act 2012 will also be engaged.

In practice, this means that AI-related risks are already regulated – albeit indirectly – through established areas of law.

Defensible AI governance for businesses

Against this backdrop, businesses would be well advised to adopt clear and defensible AI governance practices.

First, transparency is key. Where AI tools materially influence outputs or decisions that affect customers or clients, businesses should consider clearly disclosing such use in accessible policies and ensuring consistency with their terms of business. Misaligned expectations may create fertile ground for disputes.

Second, internal governance structures should be implemented to manage risk. This may include setting approval processes for the adoption of AI tools, identifying high-risk use cases, and maintaining appropriate levels of human oversight – particularly where outputs are relied on for legal, financial or operational decisions.

Third, contractual arrangements should be reviewed to ensure that risk is appropriately allocated. This includes considering how AI use is addressed in customer terms, vendor agreements and liability frameworks.

Proactive AI governance in Singapore

As AI adoption accelerates, the regulatory landscape in Asia will continue to evolve. But in the meantime, businesses operating in Singapore and the region must be aware that their use of AI comes with legal and other risks.

Proactive governance, grounded in existing legal principles, will be critical to harnessing AI’s benefits while managing its risks.

Basil Lee is an associate director and co-head of the intellectual property & TMT practice group at Helmsman in Singapore