Regulating for an AI Commons: Adopting, sustaining, and defending people-first innovation trajectories

Abstract

Technological principles of openness, interoperability, and decentralization are integral to building an AI economy that is accountable to the public and truly responsive to local needs and constraints. However, Big Tech's firm control over essential elements of the stack (such as chips, cloud, models, data, and talent) limits valuable alternatives from reaching the scale necessary for sustained growth. Worse still, hyperscalers continue to enclose and extract value from publicly funded and volunteer-driven technologies without adequately distributing these returns. We see consolidated power within a few entities and "enshittification" of once-transformative innovations. Embedding ethical micro-level tools (like open data repositories or small-language models) into workflows is promising, but without a consistent innovation strategy, they also often end up perpetuating and being appropriated by the same exploitative trajectories. Instead, this whitepaper outlines a three-pillared holistic and intentional policy approach that i) actively pushes for the adoption of cooperative technologies in existing and emerging systems, ii) sets up collective institutional structures to sustain and maintain them in the long-term, and iii) introduces regulatory proposals that limit private entities from disempowering them through uncontrolled enclosure and extraction. Through desk research and purpose-specific case studies, we argue for the importance of such an approach in the Global South, where structural constraints around infrastructure, financing, and policy governance make it difficult for AI commons and cooperative AI tools to emerge and survive. The paper concludes with key open questions for campaigners, policy advocates, researchers, and grassroots organizers to consider and explore while working towards peoples' solidarity within and between Global South countries.