Capmad assesses existing evaluations for general digital adoption and AI policy in particular. Existing assessments of global readiness need to capture African states’ progress in AI readiness fully and lay the groundwork for better utilization of evaluations in the African context.

The rapid evolution of the AI landscape in Africa

The technological landscape of artificial intelligence is undergoing rapid evolution, demanding parallel adaptation of national strategies in information technologies of African states. This case study examines the overall state of AI readiness in four African states:

• First, through the lens of existing assessment frameworks;
• Then, via a direct survey on their state of readiness.

Our initial analysis reveals that current global evaluation measures need help to capture the unique realities of African states in their journey toward AI readiness. More attention needs to be paid to national AI strategies and initiatives in these global evaluations to account for the significant work undertaken by African states to enhance their AI readiness.

From a policy perspective, we observe that while African countries face a range of challenges in harnessing their technological potential, some challenges appear to be widespread and critical:

• Weak or inadequate data protection regimes (notably in Mauritius).
• Challenges in establishing robust oversight frameworks to support sustained sectoral growth (notably in Kenya and Egypt).
• Limited governance and economic resources to dedicate to technological development and workforce (notably in Angola).

Objectively defining evaluation levels

Beyond digital readiness, several evaluations of readiness for AI policy exist. A significant part of analyzing national AI policies involves defining key readiness indicators and tracking economic and legislative responses to this emerging technology. However, in these indicators, the work of African nations has often been overlooked or excluded, even when evaluations should reflect progress and, in some cases, global leadership.

For example, in 2021, the Stanford HAI Artificial Intelligence Index Report was published, presenting a comprehensive and ongoing analysis of national AI strategies. This report mentions only two African countries that have announced AI strategies : Kenya and Tunisia. At the same time, Mauritius and Egypt had already published AI strategies in 2018 and 2019, respectively. It was in the 2023 report that this omission was corrected.

Case Studies : African States and AI Policies

In selecting our case studies, we sought states with varying maturity levels regarding national technological development, governance, and economic diversity. While many reports focus on the continent’s major economies, such as Nigeria and South Africa, or states with strong technological and innovative capacities like Rwanda’s space industry, we aimed to highlight lesser-explored states that have made efforts in recent years to enhance their infrastructure and digital capabilities.

Mauritius : Strengthening access and data protection

Despite its small size as an island nation, Mauritius is making notable strides in crafting comprehensive policies to govern its use of technology. The Mauritian government recognizes the potential of AI to enhance economic and social outcomes and transform the state into a future-ready « smart » island.

The publication of its strategic plan Digital Mauritius 2030 2018 laid the groundwork for the country’s AI readiness. This initiative marked the inception of the Mauritian Council for Artificial Intelligence. It addressed workforce skills development to meet the country’s talent needs.

This framework aims to link AI deployment to improved development outcomes and more effective governance that is better suited to citizens’ needs. Four years after the publication of the digital strategic plan in 2022, Mauritius was the top-ranked African country in the Oxford Government AI Readiness Index, with a score of 53.38 and a ranking of 57th out of 181 countries.

Egypt : Enhancing investments in AI

Egypt established a national task force early on to examine the implications of this new technology. One year after the EU’s GDPR established its process, the Egyptian National Council for AI (NCAI) was created in 2019. The Egyptian national AI strategy is built on four pillars :

• Governance
• Development
• Capacity Building
• International Relations

The strategy notably allocates a budget of 318 million USD to digital transformation projects, demonstrating the government’s commitment to advancing AI initiatives. The approach also emphasizes responsible and ethical AI practices, with plans to integrate a dedicated AI ethics stream within the NCAI.

Regarding AI readiness, Egypt is considered the second highest-ranked African country in the Oxford Government AI Readiness Index, with a score of 49.42 and a ranking of 65th out of 181 countries. The country’s weakest performance was observed in the Technology pillar, despite Egypt’s ongoing investments to enhance its ICT infrastructure and attract investments to develop its technology sector and capabilities. Egypt is also African, producing the highest number of AI research articles. It records the highest R&D expenditures on the continent.

Kenya : Creating a favorable and robust environment

The Kenyan government has proactively embraced AI as a catalyst for national development. 2018, a dedicated AI task force was established to formulate a comprehensive national AI strategy. This initiative sought to position Kenya as a global AI hub and foster widespread adoption of AI technologies in critical sectors. This task force notably laid the groundwork for the subsequent launch of the 10-year National Digital Master Plan 2022-2032 by the Ministry of Information, Communication, and Technology (ICT).

Regarding AI readiness, Kenya ranks as the sixth highest-ranked African country in the Oxford Government AI Readiness Index, scoring 40.36 and 90th out of 181 countries. Kenya’s score is bolstered by its Data and Infrastructure pillar performance. However, the low score in the Governance pillar does not reflect the government’s ongoing efforts to create an environment conducive to the practical and responsible development of AI integration.

Angola : Establishing solid foundations

While it’s noted that evaluation indicators tend to underestimate countries in areas of improvement, they more accurately identify states with low readiness scores across all levels. Angola, one of the world’s largest oil producers, ranks among the least AI-ready economies in Africa, according to Oxford Insights, placing it 38th among African countries, with a score of 24.77 and 163rd out of 181 countries.

In 2021, during the Digital Transformation Forum, the Minister of Telecommunications, Information Technologies, and Media highlighted AI’s potential to drive the country’s digital evolution. While digital transformation plans are underway, as evidenced by signing agreements with the United Arab Emirates for the Digital Angola 2024 strategy, Angola is currently focusing more on AI readiness. Thus, among all studied countries, Angola exemplifies the effectiveness of current assessments of AI readiness.

Future policy indicators

Global assessments have recently increased in measuring countries’ readiness for AI. However, when evaluating African countries’ readiness for AI, it becomes evident that current global evaluations often need to describe the nuanced landscape of these nations more adequately.

To address these challenges, global assessments must reassess their strategies for measuring indicators and data sources. Assessments need to adapt their approach to Africa’s distinct economic needs, considering the continent’s capacity, resources, and history of technological advancements.

Foundational technological infrastructure

We define foundational technological infrastructure as national policies or strategies, existing or under development, focusing on establishing essential technological engagement and deployment, notably :

• Data protection legislation.
• Development of information and communication technologies.
• Digital governance and a technically skilled workforce.

As highlighted in the Indicators section, existing sources on this topic still need to include related indicators in current assessments. To better complement these indicators, we suggest collaborating with individual countries to discover measures currently being adopted for these indicators.

For instance, in the Angola case study, we observe the priority given to including STEM education in school programs, which could be used as an indicator to enhance the technical skills of the workforce.

Economic and development context

This indicator uniquely addresses the need to consider African nations’ diverse development timelines, economic capacities, and stages of development when assessing AI readiness. Strategically recognizing challenges such as declining gross domestic product (GDP) and human development index (HDI) ensures a comprehensive understanding of Africa’s readiness for AI, which can lead to targeted solutions for specific economic contexts.

The indicator goes beyond legislative review and narrow economic indicators of existing assessments. For example, the Stanford HAI index includes three economic indicators :

• Federal budget for AI R&D.
• Defense budget requests for AI-specific research.
• Government contract spending related to AI.

In the Oxford Insights report, broader indicators such as corporate R&D spending, availability of venture capital, and the value of ICT goods and services trade exist. Different funding models or non-commercial economic power can be highlighted with the indicator as proposed here.

Regional and continental cooperation

Understanding the full potential of AI and identifying responsible development and use of these technologies is a challenge for any country. African states should have help to go it. One advantage African states have lies in the regional and continental communities they already collaborate with in other critical policy areas.

Conclusion

Through case studies of each country, Capmad identifies the limitations and gaps in current assessments of AI readiness and proposes recommendations to revise assessment methodologies to better account for the economic and political contexts of the continent. By integrating a more qualitative analysis of countries’ technology and governance strategies and initiatives, these resources constitute a necessary foundation to enhance AI readiness in Africa.