Intermediate

Global AI Regulatory Landscape

AI regulation is a global challenge with diverse approaches reflecting different values, priorities, and governance traditions. Understanding the international landscape is essential for companies operating across borders.

China's AI Regulations

China has taken a technology-specific approach, regulating individual AI applications rather than AI as a whole:

Regulation Year Scope
Algorithmic Recommendation Regulation 2022 Governs recommendation algorithms: transparency, user control, prohibition of price discrimination, protection of labor rights
Deep Synthesis Regulation 2023 Governs deepfakes and synthetic content: watermarking requirements, consent for likeness use, content review obligations
Generative AI Measures 2023 Governs generative AI services: content safety, training data legality, user registration, socialist core values alignment
AI Safety Governance Framework 2024 Non-binding guidelines on AI safety risk classification, testing, and governance across the AI lifecycle
💡
Key Difference: China's regulations are already being enforced, while many Western regulations are still in development. Chinese regulators have actively fined companies for algorithm violations and required generative AI services to register before launching.

United Kingdom

The UK has adopted a "pro-innovation" approach that avoids a single comprehensive AI law:

  • Sector-based regulation: Existing regulators (FCA, Ofcom, CMA, ICO) apply AI principles within their domains
  • Five principles: Safety, transparency, fairness, accountability, and contestability
  • AI Safety Institute: Government body for testing frontier AI models and conducting safety research
  • No new central regulator: Relies on existing regulators coordinating through a central function

Other Key Jurisdictions

Canada (AIDA)

The Artificial Intelligence and Data Act proposes requirements for high-impact AI systems, including risk assessments, bias mitigation, transparency, and a new AI Commissioner.

Brazil

Brazil's AI Bill establishes a risk-based framework similar to the EU AI Act, with requirements for high-risk systems and rights for affected individuals. Focus on protecting fundamental rights.

Japan

Japan takes a non-regulatory, guidance-based approach through its AI Strategy Council. Emphasizes "Society 5.0" integration and international cooperation on AI governance.

India

India has opted for a light-touch approach, focusing on promoting AI development through its National AI Strategy while developing sector-specific guardrails.

International Coordination

Initiative Participants Focus
OECD AI Principles 46 countries Non-binding principles for trustworthy AI, adopted by G20
G7 Hiroshima Process G7 nations International guiding principles and code of conduct for advanced AI systems
AI Safety Summits 28+ countries Frontier AI safety, testing, and international cooperation on risk management
UN Advisory Body on AI Global Developing global governance recommendations for AI
Global Partnership on AI 29 countries Multi-stakeholder initiative supporting responsible AI development
✅
Practical Implication: For companies operating globally, the EU AI Act effectively sets the floor for compliance (the "Brussels Effect"). Building to EU standards generally satisfies requirements in other jurisdictions, though China's content-specific rules add unique obligations.