The Geopolitics of Open-Source Generative AI: Freedom, Control and Power

Generative AI is emerging as a transformative technology with implications far beyond the tech sector, shaping economic competitiveness, military capabilities, and global power dynamics.

As access to its foundational building blocks (chips, models, and talent) becomes increasingly politicised, countries are deploying economic tools to shape AI diffusion. For example, the US export controls on advanced semiconductors and models reflect a strategy of maintaining technological dominance through a “small yard, high fence” approach. The idea is to impose strict restrictions on select technologies with military potential while maintaining normal economic exchange in other areas. However, recent policy measures such as the export controls indicate that the yard no longer remains small, and the effectiveness of the fence is also questionable. In response, many states, including China and the EU, are exploring open-source alternatives as a pathway to achieving strategic autonomy.

The OSI’s definition of open source AI focuses on the user's freedom to use, study, modify, and share. However, many other stakeholders often adopt looser interpretations. Some seek to "open-wash" proprietary tools as a branding tactic, while others seek to shape and own the ecosystem on which future AI applications will be developed. Thus, open source AI matters in different ways for different stakeholders.

This talk will explore how geopolitical interests are shaping the development of open-source generative AI, highlighting the tension between openness, control, and national strategy.

This talk will explore how geopolitical interests are shaping the development of open-source generative AI, highlighting the tension between openness, control, and national strategy.