A recently redesigned and updated Global Trends benchmark compared if and how artificial intelligence (AI) is being regulated across 14 jurisdictions around the world.

Scope

• Existence of AI laws or other binding frameworks addressing specified aspects of AI, as opposed to jurisdictions only relying on voluntary frameworks.
• What are the mandatory rules applicable to AI systems, where they exist. That includes prohibited practices, requirements for high-risk use cases, and obligations for generative AI models.
• Enforcement mechanisms, including enforcement bodies, applicable penalties, and significant enforcement actions already taken.
• AI industrial policy, including national strategies, sovereign AI ambitions and public funding allocations.

The covered jurisdictions in the benchmark are: Australia, Brazil, Canada, China, the EU, India, Japan, Korea, Malaysia, Singapore, South Africa, Taiwan, the UK and the US.

In a nutshell

Amongst its main findings, this benchmark showed that there are diverse approaches worldwide in regulating AI:

• four jurisdictions adopted or proposed comprehensive AI legislation (e.g. the EU);
• five others apply binding AI-related obligations through amendments to existing laws or by adopting secondary legislation to varying degrees, with China being the most extensive example;
• four jurisdictions rely solely on voluntary schemes with no binding AI-specific rules (e.g. Singapore); and
• the US has no comprehensive federal AI regulation and actively discourages state-level rules. However, several individual US states have begun adopting their own AI laws.

For more information and to access the full report, please click on “Access the full content” - or on “Request Access”, in case you are not subscribed to the Global Trends service.

See also our global benchmark on copyright and AI, as well as our detailed European benchmark on the status of national AI laws.