Artificial intelligence (AI) is emerging as a driving force for transformation in economies and societies, but West Africa faces a major challenge: establishing a governance framework capable of reconciling innovation, ethics, inclusion and sovereignty. It is around this strategic issue that the Gender and Responsible AI Network (GRAIN) brought together researchers, public policy-makers, experts, innovators and partners for a webinar dedicated to presenting its comparative report on the political, legal and institutional frameworks for AI in the Senegal, Benin, Côte d’Ivoire and Burkina Faso.
As well as presenting the findings, this meeting provided an opportunity to launch an in-depth debate on the conditions required to make artificial intelligence a genuine driver of sustainable development, social justice and economic transformation in West Africa.
AI is not just about technology
Opening the proceedings, Jean-Claude Bidogeza, Director of Research at IPAR, pointed out that artificial intelligence is now a key driver of economic and social transformation. He emphasised, however, that its development must be underpinned by sound public policies, appropriate legal frameworks and governance capable of ensuring ethics, transparency and inclusion.
He emphasised that the challenges associated with AI go far beyond the technology itself and also concern issues of governance, trust, digital sovereignty and social justice. In his view, dialogue between researchers, policymakers, the private sector and civil society is essential to developing artificial intelligence that genuinely serves sustainable development in Africa.
GRAIN: a network driving forward AI centred on GEDI principles
Dr Marame Cissé, a gender expert and thematic coordinator at IPAR, reviewed the network’s achievements since its launch in 2022. She noted that the project, led by a consortium comprising IPAR (Senegal), Sunbird AI (Uganda) and CSEA (Nigeria), with support from IDRC and FCDO, aims to promote responsible, inclusive and gender-sensitive artificial intelligence in sub-Saharan Africa. She also outlined the three main priorities for the project’s second phase: the co-development and roll-out of a GEDI (Gender, Equity, Diversity and Inclusion) framework, the analysis of public policies relating to AI, and the support of innovations in the sectors of agriculture, education and African languages.
Four countries, four different paths, but similar challenges
Presented by Fatou Fofana, a lawyer specialising in intellectual property, the comparative analysis highlights the progress made in the four countries whilst revealing the many challenges that remain to be addressed in order to strengthen AI governance in West Africa.
The analysis highlights differences in maturity levels between the four countries.
Benin stood out as the first of the four countries to adopt a national artificial intelligence strategy in 2023, followed by Senegal with its 2024–2028 strategy. Côte d’Ivoire, which adopted its AI strategy for 2030, appears to be the most advanced country in terms of integrating the principles of inclusion, ethics and data governance.
Burkina Faso, for its part, does not yet have a national strategy specifically dedicated to artificial intelligence. Nevertheless, the report highlights that it has a particularly interesting legal framework. Its law on the protection of personal data is the only one, amongst the four countries studied, to make explicit reference to processing based on artificial intelligence. The country also has a set of digital and cybersecurity policies which provide a solid foundation for the development of a future national AI strategy.
The study also shows that the four countries already have relatively well-developed frameworks for personal data protection and cybersecurity. However, none of them has yet introduced specific legislation governing artificial intelligence, nor has any established mandatory mechanisms for algorithmic transparency, auditing or redress in the event of discrimination caused by AI systems.
A mini-panel to put the results into perspective
To put these findings into context, a mini-panel, moderated by Marame Cissé, brought together Cheikh Faye, Head of Data at IPAR, Mamadou Mbodj, Senior Data Scientist at the World Data Lab, and Laetitia Badolo, Director of Advocacy and Impact at Niyel.
Mamadou Mbodj emphasised the importance of data governance, arguing that the main challenge lies not so much in the availability of data as in its accessibility, interoperability and sharing between institutions. In his view, strengthening Africa’s digital sovereignty depends above all on better governance of public data.
.For his part, Cheikh Faye emphasised the importance of better aligning future AI governance frameworks with the ethical mechanisms already in place in certain sectors, particularly healthcare. Cheikh Faye also discussed the assessment of the level of development and applications of artificial intelligence in strategic sectors, carried out in the four countries concerned. He emphasised that the results obtained are broadly consistent with those presented in the report.
He emphasised the need to put in place appropriate regulatory frameworks to address the challenges posed by emerging technologies, particularly in terms of transparency, accountability and the protection of personal data, but also with regard to data sharing, as artificial intelligence relies fundamentally on the availability and quality of data for its development.
For his part, Laetitia Badolo, Director of Advocacy and Impact at Niyel, presented the regional guidelines developed as part of the project. These guidelines offer practical guidance on how to embed the principles of inclusion, transparency, equity and accountability right from the design stage of artificial intelligence systems. In particular, they call for greater representation of women in decision-making bodies, improved data quality and the creation of a common regional framework for evaluating AI systems. They also emphasise the need to involve parliamentarians, citizens and all stakeholders in accountability mechanisms, in order to ensure democratic oversight of artificial intelligence technologies that influence life choices.
Recommendations for strengthening AI governance
Following this comparative analysis, the report sets out several recommendations designed to support West African countries in the development of responsible artificial intelligence:
- Revise national AI strategies to explicitly incorporate the principles of Gender, Equity, Diversity and Inclusion (GEDI), with measurable targets and monitoring mechanisms.
- To develop a participatory national strategy in Burkina Faso, incorporating GEDI principles from the outset.
- Establish or strengthen national authorities and coordination mechanisms dedicated to the governance of artificial intelligence.
- Enact specific legislation on AI, particularly with regard to algorithmic transparency, the accountability of stakeholders, risk management and protection against discrimination.
- Strengthen national ethical frameworks to ensure the responsible development and use of artificial intelligence.
- Improve data governance in order to reduce algorithmic bias and ensure a fairer use of data.
- To strengthen the capacity of public institutions and data protection authorities in the governance and regulation of AI.
- Invest more in training, research and innovation in the field of artificial intelligence.
- To promote women’s participation in AI-related professions and to facilitate their access to training, employment and leadership opportunities within the sector.
- Establish gender-sensitive monitoring indicators to measure progress towards inclusion and equity in AI-related policies and initiatives.
- To strengthen regional cooperation with a view to developing a harmonised framework for the governance of artificial intelligence in West Africa.
- To develop AI solutions for the agricultural sector that are accessible and tailored to the needs of smallholder farmers, taking into account local circumstances and GEDI principles.
- To incorporate GEDI principles into education policies and build teachers’ capacity to support the responsible integration of AI into education.
- To build and develop representative linguistic databases in order to better incorporate African languages into AI technologies.
- To promote African languages in artificial intelligence systems, in particular through the development of solutions tailored to local linguistic and cultural contexts.
A regional momentum to build on
The discussions confirmed that West Africa now has a strong foundation on which to build responsible artificial intelligence. However, the speakers emphasised that the progress made must be accompanied by further efforts to strengthen political, legal and institutional frameworks, improve data governance, fully integrate the principles of equity, diversity and inclusion, and accelerate cooperation between states.
Through this webinar, GRAIN reaffirms its commitment to supporting governments, researchers, innovators and civil society organisations in developing artificial intelligence that promotes sustainable development, responsible innovation and inclusion.
The findings of this study make a significant contribution to the regional debate and offer concrete courses of action for making AI a catalyst for change that benefits everyone.
On Monday 15 May 2023, IPAR took part in an "expert panel on artificial intelligence as a driver of development” in Saly Read more
IPAR took part in the second workshop on the formulation of the national Artificial Intelligence strategy, which was held on 21 Read more
FR 
EN 
