Skip to main content

Introduction

December 2 - December 4 2024 I WP3458

A,Shot,Of,A,Smiling,Female,African,Farmer,Using,A

This report covers the event “Successfully Harnessing AI in Africa” held on 2-4 December 2024 at Wiston House, Wilton Park, UK in partnership with the UK Government’s Foreign, Commonwealth and Development Office (FCDO), Google, the African Leadership University, and the African Observatory on Responsible AI.

The event convened African and international leaders from the government, academia, civil society, development partners, AI practitioners and the private sector to discuss how Africa can harness artificial intelligence (AI) to address the continent’s societal and economic challenges and drive the attainment of the Sustainable Development Goals (SDGs). The participants agreed that for AI to transform such key sectors in Africa as healthcare and agriculture requires responsible, inclusive, and context-aware AI strategies.

Throughout the event, speakers stressed the urgency of building a skilled workforce, investing in data infrastructure, making local data available for AI R&D, and creating enabling regulatory environments to ensure Africans benefit from AI. With over one-half of Sub-Saharan Africa projected to have internet access in the coming years, there is a narrow window of opportunity to ensure AI is harnessed effectively and responsibly.

Day 1

Opening remarks: setting the stage for Africa’s AI future

“There is a narrow window of opportunity to ensure AI is harnessed effectively and responsibly”

The discussions opened by highlighting the urgent need for African economies to promote AI strategies aimed at bridging the continent’s inequalities. Google introduced the new AI Policy Blueprint for Africa commissioned by Google and delivered by Nextrade Group that lays out detailed and comprehensive policies for AI development and deployment in Africa, arranged under four pillars key for AI development – infrastructure development, skills training, investment in innovation, and responsible AI policies. It was acknowledged that several efforts are already on the way to promote AI use in Africa that can be readily leveraged, such as the AI for Development program led by Canada’s International Development Research Centre (IDRC) and the UK Foreign, Commonwealth & Development Office (FCDO).

“Governments can lead by example to promote AI use”

Session 1: Exploring the transformative potential of AI for Africa

The first session examined how AI can improve availability and productivity in key sectors in Africa, including healthcare and agriculture, and in general promote the attainment of the Sustainable Development Goals (SDGs).

The session featured a healthcare innovator who described how an AI model, trained on two million real-life queries from mothers in Kenya, is providing healthcare advice to expectant mothers at scale. The AI tool answers more than 12,000 questions daily from expectant mothers at 90 percent accuracy; questions related to more serious health conditions or where the mother reports that she did not receive a sufficient response are answered by a human. The AI system provides locally contextualized, actionable information that helps reduce maternal mortality.

An AI researcher discussed how AI is transforming agriculture through disease detection models. For example, AI-powered tools enable farmers to identify cassava crop diseases early and help them increase productivity and reduce harvest losses. The application has already expanded to other crops. Access for non-English speakers can be expanded through large language models (LLMs) that have multiple languages – Google for example has added 24 African languages such as Somali and Hausa in Google Translate.

AI’s predictive capabilities are also being used in disaster management in Africa, for example to predict floods and wildfires by providing early warnings that enable timely evacuations and emergency responses.

Participants agreed that governments can lead by example to promote AI use; one example is Kenya Revenue Authority’s exploration of AI use for tax collection.

Applying and scaling these types of AI applications in Africa requires, for example, enhancing skills for Africans to apply AI; promoting access to large local datasets including government data; enhancing data privacy; and promoting AI innovation through fair use copyright laws that both protect and promote innovation. These policies are ever more important for harnessing such breakthrough technologies as agentic AI, a leading AI trend for 2025, where AI-driven agents complete complex tasks.

African countries can drive these policies through concerted national and regional AI strategies, building on ongoing work such as the African Union’s 2024 Continental AI Strategy, the UK and Canada’s AI for Development Initiative, Google’s AI Sprinters report, and the Google-supported AI Policy Blueprint for Africa delivered by Nextrade Group that stresses digital infrastructure, AI skills, and enabling policy environment as key drivers of AI in Africa.

The key recommendations for expanding the potential of AI in Africa included:

  • Promoting access to local data, including government data: Africa lacks readily available local data for building locally relevant AI models. African governments play a key role in promoting cross-border data transfer and access to government data. Universities and local institutions should be empowered to manage data-sharing agreements to advance AI research while safeguarding privacy.
  • Developing local AI expertise in Africa: African governments should scale AI training programs through partnerships with universities, research centres, accelerators, and tech companies, and offer scholarships and research grants to build a talent pipeline to innovate using AI and apply AI.
  • Expanding AI infrastructure and internet access: To scale AI use in Africa, governments need to urgently prioritize investments in digital infrastructure, including cloud services, data centres, and internet connectivity.
Session 2: Shaping regional and national AI strategies

The second session focused on building comprehensive AI strategies tailored to Africa’s needs and circumstances. Participants emphasized such strategies should help ensure that AI’s benefits are equitably distributed in Africa, African businesses own and monetize AI innovations, and African governments form part of global AI governance debates. They should focus on ways to promote enabling regulatory environments, sector-specific AI policies, and AI infrastructure and compute capabilities.

The discussion also focused on practical ways to move from AI strategies and blueprints to implementation of AI initiatives. There were five major themes.

The first was the importance of prioritizing responsible uses of AI and building trust in AI systems, by addressing data privacy violations, algorithmic bias, and economic exclusion. A policy expert noted that building trust means building representative local datasets, applying AI to Africa’s unique contexts, and incorporating African values into AI ethics, including through a regional African Charter on Ethical AI. There were also calls to move from “responsible AI” to “trustworthy AI” to bridge the trust deficits associated with AI. Kenya is leading the charge on trust in AI by hosting a Trust Summit on AI in June 2025. There are also such useful tools to track trust in AI such as the Global Index on Responsible AI co-funded by the IDRC, Global Affairs Canada, the U.S. Agency for International Development (USAID) and FCDO.

The second theme was inclusion. African AI systems need to be developed with inclusion in mind lest AI widen Africa’s existing inequalities. For example, AI applications should have innovative designs to account for persons with disabilities. Inclusion and diversity should be a starting point, not a problem to be solved ex post – and disability in the AI era should be a central theme and not be a side issue. As such, Africa’s AI strategies must also address more than technological innovation and be geared to tackling underlying socio-economic inequalities. AI can also be geared to promoting inclusion in African labour markets, by augmenting employees’ skills and productivity.

Third, the participants also explored the potential for innovative financing models to support AI development in Africa. The goal should be to enable local companies to scale AI-driven solutions while ensuring that economic value remains within the continent. Suggestions included promoting venture capital funds for African AI startups, public-private partnerships, and cross-border investments.

Fourth, to scale data available for AI, African governments should promote convergence among national data rights and privacy, consumer protection, and data transfer policies. Common, interoperable AI standards, for example in the context of the African Continental Free Trade Agreement (AfCFTA), are essential for scaling AI innovations in Africa and ensuring that African governments apply the same terminology in regulating AI deployments. AI standards should be aligned with globally emerging standards and international AI governance principles.

Fifth, turning strategies into action requires leadership at the highest level, from African heads of state. Many African AI success stories to date are due to visionary leadership. Leaders such as heads of state and finance ministries need to be sensitized to the potential of AI and its many use cases. In addition, leadership is needed to turn Africa’s many challenges to AI adoption and use into strengths – for example, the vast young population can be seen as a unique asset of developing and scaling AI applications.

The key recommendations from the session on AI strategies included:

  • Creating national and regional AI strategies: African countries should develop comprehensive AI strategies aligned with the African Union’s Continental AI strategy to promote policy coherence and African AI standards aligned with global AI governance principles. The drive toward national and regional strategies requires strong leadership from the highest levels of African governments – and awareness-building with top policy makers on AI’s potential use cases.
  • Supporting inclusive regulatory frameworks related to AI: Governments need to drive the twin goals of innovation and inclusion by crafting policies that address the trust deficit with AI and promote inclusive AI systems. This includes adopting balanced data protection policies and involving underrepresented groups in AI-related policy-making.
  • Promoting investment and financing models to catalyse AI innovation and use: African policymakers should encourage innovative funding mechanisms such as venture capital, blended financing, and impact investing to support AI startups and local entrepreneurs. AI investment hubs could be built to fund tech-enabled businesses and scale AI-driven innovations.
Closing reflections on day 1: turning potential into action

Participants agreed that Africa must lead, not follow, in defining its AI development agenda, by driving AI strategies and translating ideas into concrete policies, scaling pilot projects, and creating supportive AI ecosystems. In addition, African governments need to urgently improve infrastructures, establish skilled AI workforces, and build supportive regulatory environments.

Previous

Successfully Harnessing AI in Africa

Next

Fireside chat with Google DeepMind

Want to find out more?

Sign up to our newsletter