Countries are faced with the challenge of regulating generative AI to address issues such as privacy, intellectual property, and ethics. While existing AI regulations cover broader applications, generative AI presents unique concerns that current laws don’t fully address. This page focuses specifically on the regulation of generative AI around the world, and keeps track of any developments.
Key issues for regulating Generative AI
- Intellectual property. Many countries are still unsure how to handle the copyright of AI-generated works. There are ongoing debates over whether AI outputs can be copyrighted and who holds that right—the developer or the AI user.
- Bias and fairness. Regulators are focusing on ensuring that generative AI systems do not perpetuate biases or generate harmful content. Laws like the EU AI Act address this issue, but many countries are still behind in setting clear rules.
- Misinformation and disinformation. The use of generative AI to create deepfakes or spread false information is a growing concern. China has taken a strong stance, while other countries, including the U.S. and the EU, are developing measures to combat this risk.
Europe
- European Union. The EU AI Act is one of the first major regulations specifically targeting generative AI, a subset of general-purpose AI (GPAI). The Act places generative AI under the “high-risk” category, demanding stricter compliance. Key provisions include transparency (requiring disclosures when AI-generated content is used) and accountability for bias or misinformation.
- United Kingdom. The UK AI Regulation Framework sets out a risk-based approach to regulating AI, including generative AI.
Africa
- South Africa. In 2024, the Department of Communications and Digital Technologies published the South African National AI Policy Framework. It only addresses generative AI in the context of human oversight in critical decisions using AI, “especially generative AI”.
Asia
- China. The Cyberspace Administration of China (CAC) released the Administrative Measures for Generative Artificial Intelligence Services 1 in July 2023, which regulates the development and use of generative AI, including content generation, data protection, and algorithm transparency. The AI Safety Governance Framework references generative AI and its risks for unreliable output.
Oceania
- Australia. The National Framework for the Assurance of Artificial Intelligence in Government dedicates a section for responsible use of generative AI, listing a number of government guidelines for it use in public service.
How we can help you
- Learn more about the regulation of GenAI by attending our webinars on the topic.
- Leverage GenAI law(fully) by joining our trustworthy AI programme.
- Programme members can attend the Lexing Network conference on Generative AI for lawyers.