Singapore AI Law
AI-generated text. This page was generated using artificial intelligence.
Browse Singapore AI-related law and policy.
Canada–Singapore economic cooperation
On September 18, 2026, Canada and Singapore’s foreign ministers announced negotiations toward an economic partnership framework. Their nonbinding joint statement identifies AI research, innovation and commercialization, cybersecurity, and coordination on emerging-technology governance as areas for cooperation.[1]
See also Canada AI Law and International AI Frameworks.
AI in audit practice
On October 2, 2026, the Accounting and Corporate Regulatory Authority (ACRA) issued Audit Practice Guidance No. 1 of 2026 on responsible AI use. Auditors remain accountable for their professional judgements and audit opinions when using AI. The guidance calls for critical evaluation of outputs and underlying sources, appropriate documentation, and safeguards for client confidentiality, including understanding how tools store or reuse data. It addresses audit practice and professional responsibilities.[2]
Financial-sector AI supervision
On October 7, 2026, the Monetary Authority of Singapore (MAS) issued AI risk-management guidelines for financial institutions. The supervisory expectations cover AI, including agentic systems, and apply proportionately to institutions' risks. They address board and senior-management oversight, AI inventories and materiality assessment, lifecycle controls and third-party arrangements. Financial institutions remain responsible for risks arising from externally supplied AI. MAS said institutions should meet the expectations in sections 3–4 from October 7, 2027, and sections 5–6 by October 7, 2028.[3]
Related coverage
- International Legislation — country directory.
- Policy and Governance — comparative governance coverage.
- Insurance: Singapore AI risk-management supervision — insurer and insurance-broker coverage.
Related October 8, 2026 news
- October 8 digest: Singapore sets financial-sector AI risk expectations — event: October 7, 2026.
On-device scam detection and privacy (October 14, 2026)
Google DeepMind announced on October 14, 2026 that its collaboration with GovTech Singapore had produced an on-device scam-detection model based on Gemma. The company says the model examines conversational patterns locally and issues warnings before users disclose funds or sensitive information, allowing messaging services to add scam protection while keeping conversation analysis on the device. It describes training within a GovTech-managed environment on anonymised scam-report and intelligence datasets, with private and personally identifiable citizen information excluded, and testing against synthetic dialogues and local scam corpora. These are the partners’ announced design and governance practices.[4] See October 14 news archive.
Proposed voluntary international AI incident-reporting pilot
At the October 28, 2026 Virtual Leaders’ Meeting on Frontier AI Safety, co-hosted by Norway and Finland, Prime Minister Lawrence Wong proposed that interested countries and other parties launch a voluntary incident-reporting pilot. He suggested beginning with common incident categories and information to share, then exploring secure collection and exchange. The proposal covers learning from serious failures and near misses, including cases that cause no harm or do not trigger existing reporting duties, while respecting legitimate security and commercial sensitivities.[5]
Wong advocated practical, flexible workstreams with voluntary participation and adoption, drawing on scientific cooperation and testing practical difficulties before considering a wider framework. The speech proposes a pilot; it does not establish an operating reporting system, mandatory reporting duties or a binding international agreement.[5] See International AI Frameworks for broader cooperation and Singapore in the international directory. The original report appears in October 28 news coverage.
References
- ↑ Global Affairs Canada, joint statement on a Canada–Singapore Economic Partnership Framework, September 18, 2026.
- ↑ ACRA, “Audit Practice Guidance No. 1 of 2026: Using Artificial Intelligence Responsibly,” October 2, 2026; full guidance, 10 pages.
- ↑ MAS, “MAS Sets Out Supervisory Expectations on Responsible AI Adoption by Financial Institutions,” October 7, 2026; full official release reviewed.
- ↑ JJ Geewax and Miguel de Andrés, Google DeepMind, “Google DeepMind and Singapore Government announce new AI tool to protect citizens against scams,” October 14, 2026, full announcement including on-device processing and governance sections.
- ↑ 5.0 5.1 Prime Minister’s Office Singapore, “PM Lawrence Wong at the Virtual Leaders’ Meeting on Frontier AI Safety Co-Hosted by Norway and Finland (Oct 2026),” October 28, 2026; complete official speech reviewed October 29.