Comparing Apples to Oranges: A Taxonomy for Navigating the Global Landscape of AI Regulation

The field of AI governance is rapidly shifting from voluntary guidelines to binding regulation, leading to a complex and fragmented global legislative landscape. Issues include ambiguous definitions of 'AI regulation,' divergent frameworks that impede international collaboration, and unequal access to critical information, which increases the risk of regulatory capture. A new taxonomy is proposed to clarify the scope and substance of AI regulation, aiming to support democratic rights and enhance international alignment.

Why it matters

The rapid evolution and current fragmentation of global AI regulation pose significant challenges to international cooperation and the consistent application of democratic principles. Understanding and navigating this complex landscape is crucial for maintaining competitive advantage and ensuring responsible technological development worldwide, impacting various sectors dependent on AI innovation and deployment.

Key insights

• AI governance is transitioning rapidly from soft law, like national strategies and voluntary guidelines, to binding regulation.
• The legislative landscape is complex, with blurred definitions of 'AI regulation' leading to public misunderstanding and a false sense of security.
• Divergent regulatory frameworks pose a risk of fragmenting international cooperation efforts in AI.
• Uneven access to key information heightens the danger of regulatory capture within the AI sector.
• Clarifying the scope and substance of AI regulation is deemed vital for upholding democratic rights and aligning international AI initiatives.
• A taxonomy is introduced to map the global landscape of AI regulation, focusing on essential metrics like technology or application-focused rules and horizontal approaches.

Source

arXiv — Computers and Society — https://arxiv.org/abs/2505.13673