The world is in the midst of a race to build the most advanced and efficient artificial intelligence (AI) models, spearheaded by a handful of corporations concentrated in select countries. They are ushering in fundamental breakthroughs and are poised to capture the bulk of the value in an AI-transformed world. In this shifting landscape, policy briefs from Brazil G20-T20 have alluded to the substantial risks of value transfer from lower-income countries (consumers of AI) to creator nations. This will likely exacerbate existing inequalities. There is an urgent need to determine how value from AI advancements can be captured and distributed more equitably, in line with Sustainable Development Goal (SDG) 10 – reducing inequalities.

In this policy brief, the authors outline a set of guiding principles for achieving equitable distribution through taxation, aligned with SDG 10.4 on fiscal policies to reduce inequality. There are precedents of using taxation to further development goals such as increasing employment opportunities, stimulating infrastructure development, and enhancing education and health outcomes. In addition, scholars and practitioners have advocated for using taxation (eg, a tax on carbon emissions) to finance initiatives connected with the SDGs. The authors rely on insights from these precedents to offer guiding principles that could be used to equitably distribute the value from advances in AI, balancing the need for innovation and growth with human welfare.

The guiding principles fall into three categories: (1) identifying where and when value is being created (ie, point of taxation); (2) determining the value to be taxed; and (3) managing and redistributing the funds generated through taxation. Implementing these principles would necessitate the revision and refinement of existing tax treaties and the establishment of an international governance framework allowing for negotiations and arbitrations in a free and fair manner. Under the South African presidency, the G20 is uniquely positioned to initiate these actions to promote welfare-oriented innovation, aligned with SDG 17 (partnership for the goals), through the Task Force on Artificial Intelligence and Innovation.