← Global AI Policy Radar · Methodology
Jurisdiction Baseline · AI × Policy

AI Regulation in Singapore.

At a glance
AI-Specific Law
None yet; existing laws regulate AI
Personal Data
Privacy rules govern data-driven AI uses
Consumer Protection
AI claims must be accurate and fair
Sector Rules
Online safety and finance duties apply

Singapore does not currently have a comprehensive AI-specific law or binding, economy-wide AI rule in force. AI is regulated through existing laws, including the Personal Data Protection Act, consumer protection and online harms laws, and sector-specific rules; government tools such as the Model AI Governance Framework, which provides voluntary guidance, and AI Verify, which tests AI systems, are not legislation. Regulators can investigate, issue orders, impose financial penalties, restrict services, or bring proceedings when AI use breaches these laws.

In force today5 instruments
Personal Data Protection Act 2012 in force
The main horizontal constraint on AI using personal data. Companies must address consent or another valid basis, purpose limitation, notification, accuracy, security, retention, overseas transfers, access and correction, accountability, and breach notification across AI training, fine-tuning, inference, monitoring, and vendor use.
Read the full text ↗ · 2012
Data processingSafety & securityDisclosure
Consumer Protection (Fair Trading) Act 2003 in force
AI products and services sold to consumers must not mislead, make false claims, exploit consumers, or misrepresent capabilities, accuracy, approval, performance, or benefits. This is relevant to AI marketing, automated sales, recommender systems, chatbots, and AI-enabled goods and services.
Read the full text ↗ · 2003
Consumer protectionDisclosure
Unlock the full baseline

Get the remaining 3 instruments and the founder FAQ. Free: subscribe once and every briefing on this device unlocks.

Already subscribed? Unlock →

Online Criminal Harms Act 2023 in force
Online service providers can be subject to stop-communication, disabling, access-blocking, account-restriction, app-removal, and service-restriction directions relating to online criminal activity. AI platforms and services should have processes for responding to government directions and limiting AI-enabled scams, fraud, and other criminal content.
Read the full text ↗ · 2023
Content moderationSafety & security
Code of Practice for Online Safety: Social Media Services in force
The Code is legally binding on designated social media services, currently including major services such as Facebook, Instagram, TikTok, X, YouTube, and HardwareZone. It requires systems and processes to reduce harmful content exposure, protect children, provide reporting and user-safety tools, and support accountability, including where AI is used for generation, ranking, recommendation, or moderation.
Read the full text ↗ · 2023
Content moderationMinorsSafety & security
MAS Notices on Technology Risk Management in force
These binding notices apply to specified financial institutions, including banks, insurers, payment firms, capital-markets firms, and financial advisers. An AI system that forms part of a regulated institution's technology environment may need to meet requirements on critical systems, resilience, incident reporting, security, outsourcing, and protection of customer information.
Read the full text ↗ · 2024
PrudentialSafety & securityData processing
Also worth noting
  • Singapore's Model AI Governance Framework for Agentic AI, updated in May 2026, is voluntary guidance rather than binding law, but it is the most current official deployment framework for AI agents.
  • The Model AI Governance Framework for Generative AI and AI Verify remain important voluntary governance and testing references for developers and deployers.
  • Monitor future Singapore legislation or binding sector rules on AI safety, assurance, provenance, and high-impact automated decision-making. No comprehensive AI Act is in force today.
Pending and recent legislation0 measures

Live from the Global AI Policy Radar.

Founder FAQAI law, regulators, what applies, what is next, penalties
Is there an AI-specific law in force here?

No. Singapore has no comprehensive AI-specific statute or binding, economy-wide AI rule in force today. The Model AI Governance Frameworks, AI Verify, and related assurance tools are influential but voluntary.

Who are the key regulators for AI?

The key regulators named in the source are the Personal Data Protection Commission, the Infocomm Media Development Authority, the Competition and Consumer Commission of Singapore, and sector regulators. The Monetary Authority of Singapore is relevant for specified financial institutions. Their roles depend on the laws and sectors involved.

Which rules apply to AI systems today?

AI systems may be subject to the Personal Data Protection Act, including rules on personal data use, security, retention, overseas transfers, access, correction, accountability, and breach notification. Consumer-facing AI must comply with the Consumer Protection (Fair Trading) Act, while the Online Criminal Harms Act and the Code of Practice for Online Safety apply to relevant online services. MAS technology risk management notices also apply to specified financial institutions and their technology environments.

What is coming next, and when?

The source does not identify a forthcoming AI-specific law or provide a timeline for new rules. For now, the stated position is that Singapore relies on existing laws, sector requirements, and voluntary tools such as the Model AI Governance Frameworks and AI Verify.

What are the enforcement and penalty risks?

Enforcement can include investigations, directions, financial penalties, service restrictions, and legal proceedings when an AI deployment breaches an existing regime. Under the Online Criminal Harms Act, authorities may also direct providers to stop communications, disable or block access, restrict accounts, remove apps, or restrict services. The specific risk depends on the applicable law, sector, and conduct.

Not legal advice. For educational purposes only. AI-researched against official sources (2026-08-06), links verified.

Need a company-specific assessment? Explore Nomos →