Voice and Access in AI: Global AI Majority Participation in Artificial Intelligence Development and Governance
This white paper investigates practical remedies to increase voice in and access to AI governance and capabilities for the Global AI Majority, while addressing the security and commercial concerns of frontier AI states. We examine key barriers facing the Global AI Majority, including limited access to digital and compute infrastructure, power concentration in AI development, Anglocentric data sources, and skewed talent distributions. The paper also explores the dual-use dilemma of AI technologies and how it motivates frontier AI states to implement restrictive policies.
We evaluate a spectrum of AI development initiatives, ranging from domestic model creation to structured access to deployed models, assessing their feasibility for the Global AI Majority. To resolve governance dilemmas, we propose three key approaches: interest alignment, participatory architecture, and safety assurance.
The paper recommends near-term steps such as expanding internet connectivity, developing national data infrastructures, investing in AI education programs, and strengthening regional bodies to amplify the voice of Global AI Majority countries. It also emphasizes the importance of safety assurance mechanisms that balance security concerns with increased access to AI technologies.
By thoughtfully expanding participation while addressing potential security risks, we argue that a more equitable global AI ecosystem can be achieved. This approach ensures that AI’s immense potential benefits humanity as a whole, while the technology’s risks are collectively managed through truly global cooperation.
