It is essential to recognize that the responsible deployment of AI, its investment potential, and economic growth are not mutually exclusive.
Artificial Intelligence (AI) is a fast-moving train that everyone wants to get on, and it’s easy to see why. The numbers are enticing: globally, AI could drive up economic growth by 20%. It could add US$1.5 trillion to Africa’s economy by 2030 and create 230 million new digital jobs on the continent.
AI is already being used to predict flood patterns in Mozambique, automate tuberculosis diagnoses in South Africa, mark students’ assessments in Nigeria, and power voter registration in Kenya. In Africa’s vital but under-resourced medical industry- where there are just 18 physicians for every 100,000 people– AI is expected to boost productivity.
At this point, it is clear that AI will have an outsized impact on many important industries. However, it has become imperative for African researchers and observers to highlight the need for responsible AI governance, ensuring it does not deepen existing inequalities on the continent. AI is not the first technology to have had a major impact on Africa. While the internet, social media, and mobile phones added billions to economies and spurred new industries, they also inadvertently contributed to inequality.
For example, marginalized communities- including women and people in rural areas- are already less likely to own a mobile phone or access the internet. These groups are at risk of being left behind. Meta is facing a US$1.6 billion lawsuit for allegedly inflaming violence during Ethiopia’s civil war, which left more than 500,000 people dead. The company is also facing lawsuits for violating data protection, privacy, and local consumer protection laws in Nigeria and South Africa- decades after African users first began using its platforms.
This history raises a critical question: how do you manage high-growth technology safely, without deepening inequality?
Historically, the critical minerals used to make technology hardware have been extracted from Africa, with Africans receiving little to no added value. From the cobalt mined in the DRC that powers your iPhone to the lithium that fuels the batteries in your electronics, the continent plays a crucial role. Africa is home to 30% of the world’s critical mineral reserves– including cobalt, lithium, and graphite- all essential for manufacturing AI hardware. Yet, it captures only 10% of the global revenue generated from these resources.
Meanwhile, global tech giants- Amazon Web Services, Microsoft Azure, Google Cloud, Oracle, and Huawei- control 70% of the world’s cloud and data centre infrastructure. In contrast, despite accounting for 17% of the global population, Africa holds only 1.3% of the world’s data storage capacity. The dominance of these few players, according to New America, means they “dominate investment flows, dictate infrastructure standards, and set the terms of engagement, leaving individual governments with little leverage to shape the rules on their own.”
The risks extend to AI performance itself. A 2018 MIT and Stanford study shows that commercial AI systems misidentify darker-skinned faces at significantly higher rates- especially women- with error rates of up to 34% for Black women, compared to just 1% for light-skinned men. Most languages in the Global South are low-resourced in AI systems, and content moderation for these languages is difficult because moderators often do not understand their unique linguistic and cultural nuances.
When Global South NLP researchers develop tools and strategies to improve moderation in many low-resource languages, their work is often overlooked, underutilized or ignored by tech companies.
Economic Growth And Responsible AI Are Not Mutually Exclusive
Two camps exist: those who want to see AI in Africa beyond the lens of impact- and for good reason- and others who warn about the pitfalls of technosolutionism. However, it is essential to recognize that the responsible deployment of AI, its investment potential, and economic growth are not mutually exclusive.
It is only natural that, after years of experiencing the negative effects of outside influences and technologies, African technologists and researchers have called for safety and sovereignty to be placed at the forefront. They emphasize the need for ethical guidelines and safe use of AI, as well as the importance of creating local language models and ensuring that communities have control over their digital resources.
It is no wonder that responsible AI- a concept defined by the Global Index on Responsible AI (GIRAI) as AI principles geared towards active governance of the design, development, use, monitoring, and evaluation of AI systems for achieving peaceful and equitable human futures- has been favoured by the research community. In short, when adopting AI, it is essential that the very people it is intended to benefit are protected, and that underprivileged and marginalized groups are safeguarded by enforceable laws.
But AI laws on the continent are still in their earliest stages, and Africa lags in responsible AI readiness. Only 17 countries, including Nigeria, Kenya, South Africa, and Ghana, have developed national AI strategies. According to the 2024 Global Index on Responsible AI (GIRAI), which assessed responsible AI practices in 138 countries, African nations performed poorly. South Africa ranked highest at 42nd with a score of 27.61. Technology hubs like Nigeria, Kenya, and Ghana ranked even lower- 80th, 77th, and 86th, respectively. The index evaluated government leadership, human rights protections in AI development, and the role of civil society and non-state actors.
African researchers are in consensus on the need to champion Africa’s sovereignty and roots, and to build systems that serve Africans. Scholars like Alice Munyua and projects like the Initiative for the Development of Artificial Intelligence (IDAI) have called for funding of AI institutions and data protection bodies with real power, embedding African values such as Ubuntu, kinship economies, and feminist ethics of care into AI governance frameworks, designing algorithmic audits rooted in the realities of Africans, and ensuring meaningful participation in shaping AI narratives.
IDAI has birthed the Francophone Guidelines for Ethical, Inclusive, and Responsible AI in Africa– a set of governance guidelines that takes into consideration the context, realities, and infrastructure of Francophone West Africa. “The main goals are to support national AI strategies that reflect African realities, promote responsible data governance, and create a network of focal points to ensure implementation across Francophone countries,” Emmanuel Elolo, the president of the Internet Society branch in Togo and the principal expert behind the yet-to-be-public document, told me.
In July, Google announced a US$37 million investment to expand AI capacity in Africa, including funding for food security, local language hubs, and public sector training. But instead of the jubilation that typically follows such announcements, it was met with questions: Who will govern the AI built on the continent’s data? Who will own the compute, tools, and infrastructure?
Some African AI watchers and stakeholders expressed skepticism at the strategic positioning, reflecting growing discontent with AI projects that breed dependency and don’t allow for meaningful African participation or leadership at the implementation stage.
Kenya has the highest percentage of people aged 16+ who use ChatGPT, according to a July 2025 DataReportal survey– double that of the US and ahead of the UAE and Israel. While many celebrated the feat, others warned about the risks of cognitive debt, the innovation trap, and GPT psychosis. It’s now common to see a Kenyan HR professional complain about an increase in unedited generative AI responses, with some applicants mistakenly submitting the prompt along with their answer.
In a similar vein, just as Microsoft has expressed alignment with the European Union’s AI Act, MTN Group- which serves almost 300 million customers in Africa- has expressed its commitment to responsible AI values through its Genova initiative. The Group’s Digital Infrastructure CEO, Mazen Mroué, touted that the company is developing “AI for Africa by Africans,” a sentiment echoed in the African Union’s continental AI strategy, which emphasizes digital and data sovereignty.
MTN is joined by Africa’s largest telecom companies, including Vodafone and Orange, who are adopting a Global System for Mobile Communications Association (GSMA)- designed Responsible AI Maturity Roadmap to guide the safe deployment of AI across their operations. In the US, 46% of business executives identified responsible AI as a top objective for achieving competitive advantage.
Building an Equitable and Afro-Centric AI Future
Long-term African tech investor and researcher Stephen Deng believes Africa should focus its AI efforts on manufacturing, mining, logistics, human-in-the-loop processes, and payments- rather than becoming a “lab for social impact.” He points to AI adoption growing ten times faster than the internet, 90% of OpenAI usage coming from outside the US, and Mark Zuckerberg declaring that superintelligence is within sight. He argues that an overemphasis on AI ethics and data sovereignty could limit the continent’s ability to build practical, scalable solutions.
But I think they all go hand in hand. Yes, Africa mustn’t continue to rely solely on its raw mineral resources and large population. Instead, it needs to build infrastructure, strengthen its institutions, support entrepreneurs, and create local systems that attract respect.
At this point, many of Africa’s AI stacks will not be of African origin, due to years of advancement and being outpaced by tech giants like the U.S. But African values and creating products that are useful for Africans- even on a small scale- is still a valid and holistic strategy.
This aligns with the sentiments of Nandan Nilekani, the mind behind India’s digital revolution, who calls for a democratic approach to AI. He advocates for smaller, open-source models trained on high-quality, human-vetted data rather than massive systems under centralized control.
In a new memo posted on July 30th ahead of Meta’s quarterly earnings report- where Zuckerberg reiterated that superintelligence is in sight- one line caught my attention: “Superintelligence has the potential to begin a new era of personal empowerment, where people will have greater agency to improve the world in the directions they choose.”
Building with the realities of the continent in mind, catering to the vulnerable, creating useful products, and encouraging people outside the continent to design tools that reflect Africa’s cultural nuances are not tall orders. If guided well, AI can be a powerful tool for advancing the continent. But without intentional design and ethical oversight, it risks becoming another force that entrenches inequality. Africa must not only adopt AI- it must shape its future responsibly.
Sultan Quadri, who currently serves as senior content strategist at TechPR Africa, has six years of experience reporting on technology’s impact in sub-Saharan Africa for numerous publications including TechCabal, Semafor, and Deutsche Welle.
The views expressed here are the author’s own and do not necessarily reflect the editorial stance of The Kenyan Wall Street.