Light-touch, industry-led: The UK government previously announced that it intends to adopt a light-touch and industry-led approach, meaning that there won’t be specific legislation like the EU AI Act. However, in July 2024 government proposed a set of binding measures on AI and its intention to establish appropriate AI legislation.

Mostly self-regulated: The general regulatory approach is to foster AI innovation through the responsible use of AI. The financial regulator has issued AI-friendly guidelines and best practice for regulated firms, while the data regulator has also issued guidance for the use of personal data in AI. In May 2024, the Singapore government published a comprehensive Model AI Governance Framework for GenAI.

Currently no AI specific laws: No statutory laws on the use of AI, although the HKMA and SFC have issued guidance on the use of AI in financial services. The Privacy Commissioner has published the Artificial Intelligence: Model Personal Data Protection Framework. The Government is carrying on a consultation process on reinforcing Copyright Ordinance in the face of AI advancements.

Efforts at coordination: The OECD has updated its principles to which 49 countries are committed, including the G20 and the EU– this addresses issues related to the emergence of general purpose and generative AI. There have been a variety of international co-ordination efforts including the G7 Hiroshima AI Process; US-EU Trade and Technology Council and global AI Safety Summits.

Growing body of AI guidance and law: The US is characterized by a complex regulatory landscape which is mainly state-led. There is various sector specific regulator guidance and various state level laws with respect to AI specific activities. The US has also produced a comprehensive AI Risk Management Framework. May and September 2024 saw AI legislation in Colorado and California. Acknowledging this uniquely complex regulatory landscape, the Trump administration published in July 2025 the AI Action Plan which is a comprehensive effort aimed to solidify United States leadership in artificial intelligence.

Voluntary framework and technology-neutral regulations: AI is currently regulated through a voluntary scheme and various technology-neutral regulations such as consumer protection, online safety and privacy laws. However, the federal government has announced voluntary AI safety standards as part of its measures to regulate AI and released a proposals paper for introducing mandatory guardrails for AI in ‘high risk’ settings (Sept 2024).

Upcoming overarching AI legislation: South Korea’s AI law is aimed at supporting innovation rather than imposing restrictions. The Act on Promotion of the AI Industry and Framework for Establishing Trustworthy AI was passed in January 2025 and will take effect on 22 January 2026.

Potential change to prescriptive approach: The Government has issued non-legally binding guidelines for AI developers, providers and business users to follow compliance requirements when using AI. The current Japanese government is pushing for legislation to regulate the use of GenAI and the R&D of the AI-related technologies, which if passed, would signify a “hard law” approach.

Extensive AI regulatory regime: The EU AI Act was adopted by the EU Parliament in March 2024 and entered into force in August 2024. It is the first AI-specific regulation across the globe and has extensive regulatory and liability regime with extra-territorial reach. It will become fully applicable in two years with various provisions being implemented at different intervals.

Prescriptive and risk-based: Rules effective as of August 2023 cover all aspects of generative AI. Several rules were published to regulate other AI-related service/products (e.g. Algorithm Recommendation and Deep Synthesis Services). A comprehensive official draft AI law has been listed in the legislative agenda.