Regulatory and Ethical Frameworks for Generative AI in Singapore

Singapore’s regulatory and ethical frameworks for generative AI are centred on voluntary, adaptable governance guidelines rather than prescriptive legislation, with a strong emphasis on responsible, transparent, and human-centric AI deployment.

Key elements include:

• Model AI Governance Framework for Generative AI (Model Gen-AI Framework): Published in May 2024, this framework extends Singapore’s existing Model AI Governance Framework to address generative AI specifically. It sets out nine key dimensions for responsible AI use, including accountability, data quality, trusted development, incident reporting, testing and assurance, security, content provenance, safety and alignment research, and AI for public good. This framework is voluntary and designed to be adaptable to evolving technology and use cases.

• Voluntary and sector-specific approach: Singapore does not have dedicated AI laws but relies on existing legislation (e.g., data protection under the Personal Data Protection Act, health products regulation for AI-enabled medical devices, road traffic laws for autonomous vehicles) supplemented by guidelines. This approach allows flexibility and careful study of risks before considering permanent legislative changes.

• Regulatory sandboxes and testing tools: Singapore launched initiatives like Project Moonshot and the Generative AI Evaluation Sandbox to provide controlled environments for testing large language models and generative AI systems. These tools help address safety, security, and ethical challenges while fostering innovation.

• Ethical principles: The governance framework emphasises explainability, transparency, fairness, and human-centric AI. Organisations are encouraged to implement internal oversight, stakeholder communication, and risk management aligned with these principles.

• National AI Strategy (NAIS 2.0): Singapore’s updated AI strategy focuses on excellence in AI advancement and confident AI adoption, supporting talent development, infrastructure, data accessibility, and a pro-innovation regulatory environment with appropriate guardrails. It also promotes international collaboration and positions Singapore as a hub for AI innovation and governance.

• Agentic AI and emerging technologies: While no specific laws regulate agentic AI, Singapore’s AI Practice Group published an Agentic AI Primer in 2025 to guide autonomous AI systems, especially in the public sector. Enforcement of AI-related issues currently falls under existing laws on data protection, cybersecurity, copyright, and online safety.

• Sectoral applications: The Model AI Governance Framework has been successfully applied in finance and healthcare to ensure fairness, transparency, and bias mitigation, such as in credit scoring and AI diagnostic tools for diabetic retinopathy.

In summary, Singapore’s approach is nimble, flexible, and collaborative, focusing on voluntary frameworks and guidelines that evolve with technology, supported by testing environments and international cooperation, rather than rigid AI-specific legislation. This strategy aims to balance innovation with ethical safeguards and public trust.