The advent of artificial intelligence (AI) has the potential to transform African cities, but its successful integration requires a comprehensive understanding of its implications and the development of appropriate governance frameworks. The digital sector is set to grow by around $180 billion by 2030, with jobs in AI expected to grow by over 200% by 2025. AI has the potential to address challenges such as poverty, inequality, and limited access to essential services, such as improving healthcare outcomes, enhancing agricultural productivity, and optimizing urban infrastructure. AI governance, where AI ethics meet implementation through institutions and regulation, is crucial for upholding these values in the AI ecosystem. The Global Index on Responsible AI provides governments with baseline information on their country's AI activities, offering clear benchmarks for countries to ensure their citizens' rights are protected and promoted in the context of AI use. Responsible AI refers to the design, development, deployment, and governance of AI in a way that respects and protects all human rights and upholds the principles of AI ethics.
“AI governance, the way in which I like to define and describe it, is where AI ethics meets implementation through institutions and regulation. So AI ethics are the values that should guide the use and development of artificial intelligence.”
The advent of artificial intelligence (AI) has ushered in a new era of technological advancement, reshaping industries and societies worldwide. As AI continues to evolve, its potential to transform African cities is immense. However, the successful integration of AI requires a comprehensive understanding of its implications, as well as the development of appropriate governance frameworks.
The digital sector is primed for immense growth, accounting for around $180 billion by 2030, set to increase as a consequence of the enactment of the African Continental Free Trade Area (AfCFTA). Jobs in AI are likely to grow by over 200% by 2025. AI’s potential is already being used to help detect cases of tuberculosis in Nigeria, combatting poachers in Malawi, and to prevent crop disease by Cashew farmers in Ghana. Further, a program run by the South African Department of Health, called ‘MomConnect’ uses AI chatbots and software to assist over 1.8 million mothers with pre- and post-natal care.
AI can be broadly defined as the simulation of human intelligence in machines, enabling them to perform tasks that typically require human cognition. This includes learning, reasoning, problem-solving, perception, and language understanding. AI applications are diverse, ranging from natural language processing and computer vision to machine learning and robotics.
In the African context, AI has the potential to address a wide range of challenges, including poverty, inequality, and limited access to essential services. For instance, AI-powered systems can be used to improve healthcare outcomes, enhance agricultural productivity, and optimize urban infrastructure.
AI governance really should be about how we create rules and standards that can guide everybody within the AI ecosystem to ensure that we uphold these values as AI is designed, developed, used, governed and so forth. The Global Index on Responsible AI (GIRAI) is the first tool to set globally-relevant benchmarks for responsible AI and assess them in countries around the world. This study constitutes the largest global data collection on responsible AI to-date. In its first edition, the Global Index on Responsible AI covers 138 countries and jurisdictions, including 41 countries from Africa.
“One of the most significant things about the Global Index on Responsible AI is it included data collected from 41 African countries and no study had ever been so comprehensive in assessing responsible AI or even just AI across so many African countries.”
One of the key implications of the findings of the Global Index on Responsible AI for Africa is the need for increased investment in local AI capacity. By fostering a vibrant AI ecosystem, African countries can develop their own AI technologies and applications, reducing their dependence on foreign solutions. Additionally, the index highlights the importance of building a skilled workforce capable of harnessing the power of AI.
Rachel stated that the Global Index on Responsible AI provides governments with baseline information on their country's responsible AI activities and commitments. She emphasized that AI is often multi-sectoral and requires a comprehensive understanding of government-wide efforts. The index offers clear benchmarks for countries to ensure that their citizens' rights are protected and promoted in the context of AI use. These benchmarks include guidelines on public procurement, building AI skills, implementing AI ethics principles, and having data protection laws and institutions. By conducting the index annually, governments can track their progress in AI adoption and its impact on their societies.
“The Global Index on Responsible AI showed us that just because you have a policy in place does not mean that you are ensuring that these technologies are governed in a way that is ethical, responsible, and protects human rights and democracy. So having a Policy alone - and policies are not binding - they just kind of demonstrate executive intent. They demonstrate that a government cares about this issue and recognizes its importance and will do a set of things.”
Rachel continued to further state that they have a problem on the continent where they develop policies but lack the funding or capacity to implement them. She emphasized the need for active agencies within countries, such as oversight institutions, data protection authorities, human rights commissions, and sectoral level bodies, to understand the impact of AI on their sectors and adjust accordingly to ensure that societal values are not negatively impacted. Additionally, she highlighted the importance of networks of people, particularly civil society organizations, to be involved in this process.
She continued by emphasizing the important role of universities and research communities in understanding the best ways to govern AI. She highlighted the lack of knowledge about the precise impacts of AI on African societies and economies, emphasizing the need for research to inform policy decisions. She also cautioned against implementing rigid government approaches similar to those in the European Union, which could hinder the growth of local innovators. Instead, she advocated for governance mechanisms that support the growth of local industries and do not impose excessive compliance costs. Additionally, Rachel stressed the importance of civil society in advocating for change, educating the public about AI, and ensuring that people's rights are protected.
“The responsible design, deployment and governance of AI are proportionate to the purpose of its use and meet the technological needs of the individuals and societies it seeks to serve.”
When it comes to tracking and measuring responsible AI, it’s important that it’s clearly defined. That means defining responsible AI through a human rights lens. Responsible AI refers to the design, development, deployment and governance of AI in a way that respects and protects all human rights and upholds the principles of AI ethics through every stage of the AI lifecycle and value chain. It requires all actors involved in the national AI ecosystem to take responsibility for the human, social and environmental impacts of their decisions. Rachel mentioned that South Africa had a low score for frameworks, indicating a lack of governance efforts. However, South Africa scored highest in government actions, demonstrating active involvement in AI development and safeguarding. Other countries with interesting AI policies include Senegal and Egypt, which focus on building local AI ecosystems and promoting international cooperation. While these countries have made progress in AI governance, the implementation of regulatory standards is still awaited. Rachel also mentioned her involvement in drafting the African Union Continental AI strategy, which aims to establish such standards for the continent.
“The African Observatory on Responsible AI serves as a valuable platform for knowledge sharing and collaboration on AI-related issues.”
It brings together researchers, policymakers, and industry experts to discuss the potential of AI for the continent and address the challenges associated with its adoption. The African Observatory plays a crucial role in promoting the development of AI governance frameworks. By fostering dialogue and consensus among stakeholders, the Observatory can help to ensure that AI is used ethically and responsibly in Africa.
Rachel mentioned that the African Observatory on Responsible AI is working to build the capacity of researchers across the continent in AI ethics. The Observatory is launching two certificate courses, one on AI and human rights in Africa with the University of Pretoria and another on African AI ethics and policy with the University of Cape Town. Additionally, they are working to raise the visibility of African voices and expertise in global AI debates and decision-making.
“AI can play a significant role in addressing challenges related to Africa rapid urbanization and improving the quality of life for urban residents.”
In terms of urban Infrastructure, AI can optimize traffic management, energy consumption, and waste collection. For instance, in Nairobi, AI-driven traffic management systems help reduce congestion by adjusting traffic light patterns in real-time. Similarly, in Cape Town, AI is used to monitor and manage energy consumption, leading to more efficient use of resources.
In healthcare, AI-powered diagnostic tools and personalized medicine can improve healthcare access and outcomes. In Kenya, chatbots like SophieBot provide medical advice and information, making healthcare more accessible, especially in remote areas. In Nigeria, AI-driven platforms like Ubenwa use machine learning to detect birth asphyxia from a baby’s cry, potentially saving lives.
In education, AI can personalize learning experiences and enhance educational outcomes. In South Africa, the Siyavula platform uses AI to provide personalized learning experiences in mathematics and science, helping students improve their understanding and performance.
In financial Services, AI can enable more inclusive financial services and improve access to credit. In Ghana, AI-powered microlending apps like Fido use machine learning to assess creditworthiness and provide loans to individuals who might not have access to traditional banking services.
“While the potential of AI for African cities is immense, there are several challenges that need to be addressed.”
The lack of high-quality data can hinder the development and deployment of AI applications. Rachel stated that more data is needed and that existing data should be used more effectively. She emphasized the importance of data governance and data protection, noting that most African countries have data protection laws and authorities. She advocated for implementing these laws, capacitating oversight institutions, and informing people about their rights. Additionally, she highlighted the need to ensure responsible data collection practices and create a culture of data protection compliance and respect for personal data and emphasized that existing laws can address these issues rather than requiring new ones.
Infrastructure limitations, such as insufficient internet connectivity and computing power, can constrain AI adoption. Additionally, talent shortages due to the scarcity of AI skills and expertise can limit the development and implementation of AI solutions. Ethical concerns also arise with the use of AI, particularly related to privacy, bias, and accountability. Rachel acknowledged that AI systems are often biased due to their reliance on Western data and worldviews. She emphasized the need to develop more representative data sets from African sources and to prioritize African-generated content in training these systems. Furthermore, she highlighted the importance of ensuring a fair and ethical supply chain for content and data generation, including intellectual property issues and worker protection. Adams stressed the importance of considering these factors when building capacities for more representative AI systems.
Rachel Adams also expressed deep concern about the worsening of global inequality due to AI. She pointed out the bias and elitist nature of AI, which disproportionately benefits those with access to technology. To address these imbalances, Adams emphasized the need to ensure that AI does not exacerbate existing inequalities. She also voiced her concern about the lack of focus on structural inequality in AI research and governance. Adams underscored the critical role of governance in understanding and addressing these inequalities, particularly emphasizing the efforts of the Global Center on AI Governance.
On AI colonization, Rachel acknowledged the question and mentioned her upcoming book, "The New Empire of AI: The Future of Global Inequality," which addresses the issue of AI colonization. She emphasized the need for regulations and guidelines to prevent exploitation by foreign AI firms and the importance of building local AI ecosystems. While acknowledging the presence of AI, she suggested that Africa should focus on building its own capacity to use AI for its own challenges rather than rejecting it entirely. She highlighted the existing talent and innovative AI applications in Africa that uphold African values and encouraged their support and promotion for a positive impact on African societies.
“To overcome these challenges, African countries need to invest in data infrastructure, develop AI talent, and establish ethical guidelines for AI development and use. Additionally, international cooperation and knowledge sharing can play a crucial role in accelerating AI adoption in Africa.”
Rachel continued to state that governments should primarily focus on regulating the emerging AI industry. She suggested that governments share their data with trusted partners through well-crafted public procurement or data sharing guidelines. While governments don't need to build AI themselves, they can share resources to enable locally appropriate AI development.
She emphasized that AI should not be used as a solution for everything and highlighted the importance of considering when AI is the right solution and when it is not. Adams recommended that governments upskill their public servants and policymakers to make informed decisions about AI usage, ensuring ethical and responsible implementation.
The Global Index on Responsible AI and the African Observatory on Responsible AI provide valuable insights into the potential of AI for African cities. By addressing the challenges and seizing the opportunities, African countries can harness the power of AI to improve the lives of their citizens and drive sustainable development.
