AI: Six African countries want common rules
Senegal, Burkina Faso, Côte d’Ivoire, Mali, Guinea and Benin have adopted joint guidelines to provide a framework for the development of artificial intelligence. The aim is to promote responsible AI that is tailored to African realities and less reliant on rules devised elsewhere. The challenge now is to translate these principles into concrete measures.
- Unveiled on 6 July 2026, this common framework covers ethics, the protection of rights, data governance, inclusion and digital sovereignty. It is intended to help the six countries shape their public policies and coordinate their positions in international discussions
- However, it is not a common law. The document serves as a framework for guidance: each country will have to decide how to incorporate its recommendations into its own strategies, institutions and, where necessary, legislation.
Six countries to define their own approach to AI
Artificial intelligence is making headway in African businesses, public administrations and financial services. But its adoption also raises questions: who controls the data used to train the models? How can algorithmic discrimination be prevented? And how can we ensure that the benefits of these technologies also reach local communities? It is in light of these issues that Burkina Faso, Senegal, Côte d’Ivoire, Mali, Guinea and Benin have drawn up a framework entitled ‘Guidelines for Ethical, Inclusive and Responsible Artificial Intelligence in Francophone Africa’. The document was presented on 6 July 2026, on the sidelines of the first United Nations Global Dialogue on AI Governance. The initiative was led by the Burkinabe Ministry for Digital Transition, with the support of the advocacy organisation Niyel. The African Union Development Agency (AUDA-NEPAD) and the Burkinabe research centre CITADEL also contributed to the work.
Common principles, from transparency to digital sovereignty
The framework is based on seven fundamental principles and five strategic priorities. In particular, it emphasises respect for human rights, transparency, accountability, equity, inclusion, and cultural and linguistic diversity. These principles address very specific concerns. An AI system used to assess credit applications, select job candidates or refer patients may replicate biases present in its training data. If this data fails to accurately represent certain populations or languages, the results may put them at a disadvantage. The language issue is particularly important in French-speaking Africa. National languages remain under-represented compared with the dominant languages in many digital tools. Developing systems that are better suited to local contexts therefore requires access to relevant data, technical expertise and research resources. The document also addresses data governance and digital sovereignty. Behind these terms lies a strategic question: how can African countries retain the power to make decisions about the technologies they use, their data and the infrastructure on which they depend?
A regional framework that does not replace national strategies
The six countries are not all starting from the same point. Some already have national strategies dedicated to artificial intelligence, whilst others are still working on shaping their policy in this area. In Senegal, the national digital strategy forms part of the Technological New Deal launched in February 2025. Côte d’Ivoire has also embarked on its own initiative in the field of AI. In Guinea, a national roadmap covering the period 2026–2035 was approved in December 2025. The regional framework does not, therefore, replace these policies. Rather, it seeks to establish common benchmarks, facilitate cooperation and provide countries with a shared basis for their regulatory work. Such coordination can also be valuable in international negotiations. When standards relating to AI are discussed in multilateral forums, coordinated positions can help states to better defend their priorities, particularly in relation to data, languages and economic development.
The real challenge begins after publication
A joint document alone is not enough to regulate the use of artificial intelligence. For its recommendations to have any effect, they must be incorporated into public policy, backed by adequate funding and implemented by institutions with the necessary expertise. It will also be necessary to provide training for government departments, raise awareness amongst businesses and build the capacity of local researchers and developers. Without these resources, the principles risk remaining mere declarations of intent. The debate is also progressing in the parliamentary arena. In September 2026, the National Assembly of Senegal, in collaboration with Niyel, organised a workshop on the regulation of artificial intelligence, bringing together participants from Senegal, Benin and Burkina Faso, amongst others. Discussions centred on existing legal frameworks, gaps to be addressed and ways to strengthen cooperation between parliaments. This approach shows that the issue is no longer one that concerns only digital specialists. It also affects legislators, public authorities and citizens, whose rights may be affected by automated decisions.
An African ambition that will have to be put to the test
Africa already has a continental strategy on artificial intelligence, adopted by the African Union in 2024. The framework developed by the six countries complements this initiative at a more targeted regional level. Its relevance will now depend on its ability to influence national decisions: new regulations, monitoring mechanisms, training programmes, local research and investment in infrastructure. For Senegal and its five partners, the challenge is not simply to adopt tools designed elsewhere, but to play a part in defining the conditions under which AI will be developed and used. The common framework is now in place. The next step will be to see which countries actually take it on board, what measures they adopt and how they report on them.
