What Might a Just AI Global Governance Framework Look Like for Africa?

Research ICT Africa

The United Nations Secretary-General has called for globally coordinated action on AI governance as the only way “to harness AI for humanity while addressing its risks and uncertainties as AI-related services, algorithms, computing capacity and expertise become more widespread internationally”. The Tech Envoy has elaborated on the structures and processes required to achieve this. On the basis of the UN Common Agenda and the commitment to a Global Digital Compact, he has identified the underlying issues as the expanding digital and data, governance, innovation, and value divides. He has proposed a 32-person multistakeholder and multidisciplinary AI Advisory Panel to chart a way ahead on the global governance of AI in the remaining 12 months leading up to the Summit of the Future. This call for submissions is in support of this process. 

Research ICT Africa (RIA), an African digital policy and regulatory think tank, welcomes the opportunity to contribute to ensuring that the long-standing underlying inequalities that have produced not only the uneven distribution of harms but, importantly, uneven opportunities between and within countries to create value, are redressed. Building a Common Framework for AI that will ensure the transparent, safe and fair design and use of AI that is under human control is essential to deal with the inevitable growth of data-driven technologies. Equally important is ensuring governance that is transparent and accountable at the core, and which combines existing “international guidelines and norms, regulatory frameworks and technical standards [and] human values and societal goals”. 

Spotlighting AI as the latest challenges of global governance and simply reiterating the rhetoric of the past 30 years on the divides that need to be addressed, however, will not fundamentally transform the currently asymmetrical outcomes of digital development. Without a renewed plan of action to ensure better progress in reaching the lapsed ICT targets behind the Sustainable Development Goals (SDGs) there will be no basic public data in large parts of the majority world to serve as a useful index of measurement, or for evidence-based policy more generally. Nor will the far greater demand-side human development challenges that RIA’s research now shows are the primary barriers to redressing digital inequality be addressed; as resources and attention are once again diverted from the incomplete foundational project of universal access and digital transformation. It is crucial that, with the global challenges of governing the fast-moving terrain of AI, resources are put into developing alternative regulatory and governance strategies to ensure more equitable and just policy outcomes. Further, important as they are, transparency and accountability governance frameworks for AI will not suffice to shift the unjust outcomes of current AI business practices and self-regulation. Separating the inequalities and injustices perpetrated in the automated decision-making in millions of transactions each day perpetuates the status quo – negatively impacting those at the intersections of multiple inequalities. 

Abdicating the development of citizen-enabling digital governance frameworks to the globally dominant providers of those services enables the maintenance of the uneven power relations, and domination of knowledge systems and public resources by a handful of global tech monopolies and their governments.

From a broad African vantage point, it is important that the UN ensures participation across the spectrum of governance (from norms and principles, through to regulations and their implementation), and across the different sites (public, private, and UN). It is also vital that such participation is on a multistakeholder basis, representing the breadth of interests and expertise in Africa concerning the significance of the development and deployment of AI within technical, human rights, governance and other parameters present in various forms around the continent. Drawing on our submission to the Global Digital Compact and partnership with the UN Economic Commission for Africa (ECA) on the African review of the Compact we submit that the overarching question for the global governance of AI is what underlying policies and forms of governance are required to realise digital public goods at the national level to more equitably harness AI for the good of all.