G20 backs ‘AI Prosperity Compact’ as global rules on artificial intelligence take shape

At their innovation ministerial meeting, G20 countries endorsed a set of “AI Prosperity Objectives” and an “AI Prosperity Compact,” signalling a more coordinated push to shape global rules for artificial intelligence. Agreed in a statement dated 2 September, the compact is designed as a shared framework for using AI to drive growth and address social challenges, while limiting human-rights abuses and national‑security risks.

The statement highlights several priorities: building a skilled AI workforce, investing in digital and computing infrastructure, and developing international technical standards that govern how AI systems are designed, deployed and audited. It also stresses “responsible AI” governance, including transparency, safety testing and human oversight for powerful models such as generative AI.

For Japan and other export‑oriented economies, the G20 move matters because it will influence how domestic rules line up with the EU’s AI Act and emerging national safety guidelines. Industry groups are already warning that poorly designed regulations could hit small firms and startups hardest by raising compliance costs. The real test will be whether governments can coordinate on guardrails without choking off open innovation and cross‑border data flows that AI companies rely on.