INTRODUCTION

The inaugural installment of this piece was necessarily foundational, providing the background to the emergence of AI; how it transformed the digital space; applicable regulatory frameworks; its algorithimic transparency/accountability; its ethical dimensions and implications and the threat of foreign tech dominance/digital colonialism. This sophomore edition traces the evolution of AI; notable developments; the history of technological dependency in Africa and policy trends in the continent and beyond. Enjoy.

THE EVOLUTION OF AI

AI has progressed from rule-based systems to machine learning and deep learning models capable of autonomous decision-making. Applications range from healthcare diagnostics to autonomous vehicles, predictive policing, and financial algorithms. While AI enhances productivity, concerns arise over:

– Job displacement due to automation. (Brynjolfsson, E., & McAfee, A. (2014). The Second Machine Age: Work, Progress, and Prosperity in a Time of Brilliant Technologies. W. W. Norton & Company).

– Surveillance capitalism, where personal data is exploited for profit.

– Algorithmic governance, where AI influences public policy without sufficient oversight (O’Neil, C. (2016). Weapons of Math Destruction. Crown Publishing.). The conceptual origins of Artificial Intelligence (AI) can be traced to the mid-20th century, when pioneering figures such as Alan Turing and John McCarthy began to explore the possibility of creating machines capable of simulating human intelligence. Turing’s seminal 1950 paper, “Computing Machinery and Intelligence,” posed the provocative question, “Can machines think?”—a question that laid the philosophical groundwork for modern AI research. (Turing, Alan M. “Computing Machinery and Intelligence.” Mind 59, no. 236 (1950): 433–460.) McCarthy, who coined the term “artificial intelligence” in 1956, convened the historic Dartmouth Conference, widely considered the birth of AI as a formal field of inquiry (McCarthy, John et al., “A Proposal for the Dartmouth Summer Research Project on Artificial Intelligence,” (1955).

EARLY ASPIRATIONS AND TECHNOLOGICAL MILESTONES

Early AI efforts focused on symbolic logic, rule-based systems, and expert systems, which relied on hand-coded rules to simulate decision-making processes. These systems, while limited in scope, found application in fields such as medical diagnostics (e.g., MYCIN) and chess-playing algorithms. The emergence of machine learning in the late 20th century—particularly supervised learning techniques—ushered in a new era in which machines could learn patterns from data rather than rely solely on pre-programmed rules.

The exponential growths in computing power, availability of big data, and algorithmic innovation have since culminated in what many scholars refer to as the “AI revolution.”

NOTABLE DEVELOPMENTS

Notable developments include deep learning techniques powered by artificial neural networks, natural language processing exemplified by large language models (LLMs), and computer vision systems that rival or exceed human performance in specific domains (LeCun, Yann, Bengio, Yoshua, and Hinton, Geoffrey. “Deep Learning.” Nature 521, no. 7553 (2015): 436–444).

                              FROM AUTOMATION TO AUTONOMY

AI has transitioned from automating repetitive tasks to performing complex cognitive functions previously thought to be the exclusive domain of humans. Self-driving cars, AI legal assistants, autonomous drones, and AI-generated art demonstrate the breadth of AI’s applications. As these systems grow in sophistication, they increasingly exhibit autonomy—the capacity to make decisions and take actions without direct human intervention. This shift raises profound questions about accountability, transparency, and control.

ACCOUNTABILITY, TRANSPERENCY AND CONTROL

For example, autonomous weapons systems capable of selecting and engaging targets without human oversight challenge existing norms under international humanitarian law (IHL). Similarly, AI systems deployed in judicial or parole decisions raise concerns about bias, fairness, and due process, especially when the logic behind decisions is opaque even to their developers—a phenomenon referred to as the “black box problem.” (Pasquale, Frank. The Black Box Society: The Secret Algorithms That Control Money and Information. Harvard University Press, 2015).

HISTORY OF TECHNOLOGICAL DEPENDENCY IN AFRICA

The critique that digital technologies embody, enable, or reproduce colonial power relations is not new. As early as the 1970s, debates around communication and technology were linked to questions of sovereignty, inequality and dependency. In March 1976, at the Non-Aligned Media Seminar in Tunis, representatives from 38 NAM states and 13 observers declared that “colonialist, imperialist and racist powers have created effective means of information and communication which are conditioning the masses to the interests of these powers.” This seminar built on earlier efforts of the Non-Aligned Movement (est. 1955), which, by its 1973 Algiers summit, had embraced the decolonization of information, communication, and culture as part of the wider struggle for independence.

The Tunis meeting marked the birth of the New World Information and Communication Order (NWICO); a call to redress global inequalities in media ownership, information flows, and infrastructure. Tunisian minister Mustapha Masmoudi highlighted the imbalance: “Almost 80 percent of the world news flow emanates from the major transnational agencies; however these devote only 20 to 30 percent of news coverage to the developing countries”. NWICO gained traction at UNESCO, culminating in the 1980 MacBride Report, which directly challenged the Western doctrine of “free flow of information.” The United States and the UK eventually withdrew from UNESCO in protest, but NWICO left a lasting intellectual and political legacy: it framed global communication as a site of structural inequality and technological dependency.

Building on these debates, communication scholars introduced the idea of electronic colonialism. Herbert Schiller’s Mass Communication and American Empire argued that U.S. commercial media systems were becoming instruments of empire. Thomas McPhail later extended this, defining electronic colonialism as “the dependent relationship of poorer regions on post-industrial nations, caused and established by the importation of communication hardware and foreign-produced software, along with engineers, technicians and related information protocols” (Jacob Mahlangu, ‘Technological Apartheid: The Digital Divide Between Africa and the West’ (Sagepub.com, 6^th May, 2025) < https://advance.sagepub.com/doi/full/10.31124/advance.174652029.93416488/v1#:~:text=The%20digital%20divide%20between%20Africa%20and%20the%20West%20represents%20not,colonialism%2C%20and%20contemporary%20digital%20imperialism). This lens made clear that dependency was not only economic but also infrastructural and epistemic.

Parallel critiques arose in anthropology and development studies. Post-development theorists such as Arturo Escobar and James Ferguson argued that development projects often failed to empower but instead re-entrenched colonial hierarchies. They identified technology as a key tool in this process, framed as a “solution” but often deployed in paternalistic ways that deepened dependency. ICT4D (Information and Communication Technologies for Development) initiatives of the late 1990s and early 2000s exemplified this tension. While promising to democratize knowledge and spur development, many projects replicated older patterns: reliance on imported technology, disregard for local context, and reinforcement of global asymmetries.

By the late 2000s, scholars in postcolonial computing extended these critiques to human–computer interaction (HCI). They demonstrated how design practices in “development tech” mirrored colonial flows: low-cost labor and raw materials from the Global South, transformed into finished products exported back under narratives of benevolence. The One Laptop Per Child (OLPC) project (< https://laptop.org/ > Accessed on 16^th September, 2025) epitomized this, marketed as a humanitarian innovation but dependent on the feminized labour of Asian workers in global supply chains.

In 2013, Dal Yong Jin introduced platform imperialism, analyzing how U.S. tech giants like Google, Apple and Facebook exerted global dominance through platform capitalism, intellectual property regimes and cross-border expansion (Jin, Dal Yong, ‘“The Construction of Platform Imperialism in the Globalization Era.” Triple C: Communication, Capitalism & Critique. Open Access Journal For a Global Sustainable Information Society’ ( Researchgate.net, January, 2013) < https://researchgate.net/publication/275652379_Jin_Dal_Yong_2013_The_Construction_of_Platform_Imperialism_in_the_Globalization_Era_Triple_C_Communication_Capitalism_Critique_Open_Access_Journal_For_a_Global_Sustainable_Information_Society_111_145-#:~:text=Abstract,accumulation%20in%20the%20digital%20age. > Accessed on 16^th September, 2025). His argument made explicit that digital platforms were not neutral infrastructures but instruments of geopolitical power.

These intellectual trajectories resonate strongly with dependency theory, advanced by Samir Amin, which argued that underdevelopment in the Global South is not accidental but structurally produced through dependence on the North. Applied to technology, this means Africa’s reliance on imported hardware, software, and infrastructures reinforces systemic subordination in the global digital hierarchy. Postcolonial thinkers like Frantz Fanon and Edward Said similarly highlighted how colonialism survives in cultural, psychological, and technological forms, keeping the Global South positioned as consumer rather than producer.

From NWICO to electronic colonialism, from ICT4D critiques to postcolonial computing and platform imperialism, the throughline is clear: each era has witnessed renewed forms of technological dependency. What changes are the technologies themselves: satellites, mass media, ICTs, platforms, and now AI, but the structural critique persists. Today’s debates on digital colonialism continue this intellectual lineage, reframing old concerns around sovereignty, extraction and dependency in terms of data, algorithms and artificial intelligence. Far from a rupture, this is the latest chapter in a long struggle for technological self-determination in Africa and the wider Global South.

AI POLICY TRENDS GLOBALLY AND IN AFRICA

Global AI policy is crystallizing around a few core themes: risk-based regulation of high-impact systems, the embedding of human rights (< https://2021-2025.state.gov/risk-management-profile-for-ai-and-human-rights/#:~:text=In%20March%202024%2C%20all%20193,the%20enjoyment%20of%20human%20rights.%E2%80%9D > (State.gov, 25^th July, 2024) Accessed on 10^th September, 2025) and ethics principles, and the development of technical standards to operationalize trustworthiness. The European Union’s AI (https://digital-strategy.ec.europa.eu/en/policies/regulatory-framework-ai > Accessed on 10^th September, 2025) Act illustrates this risk-based approach by classifying systems according to potential harm and imposing proportionate obligations, while still promoting innovation. Similarly, the OECD AI Principles (< https://oecd.ai/en/ai-principles > (OECD.ai) Accessed on 10^th September, 2025), the NIST AI Risk Management Framework (US) (< www.nist.gov/itl/ai-risk-management-framework > (NIST.gov) Accessed on 10^th September, 2025), and UNESCO’s global AI ethics recommendations (< www.unesco.org/en/artificial-intelligence/recommendation-ethics#:~:text=Recommendation%20on%20the%20Ethics%20of%20Artificial%20Intelligence,human%20oversight%20of%20AI%20systems. > (UNESCO.Org) Accessed on 10^th September, 2025) provide international benchmarks centered on transparency, accountability, robustness, and human oversight.

USA, EU, CHINA’S PREFERENCES

National strategies, however, diverge. The United States favours voluntary, sector-specific frameworks to preserve innovation flexibility (Tatevik Davtyan, ‘THE U.S. APPROACH TO AI REGULATION: FEDERAL LAWS, POLICIES, AND STRATEGIES EXPLAINED’ (scholarlycommons.law.case.edu, 24^th January, 2025) < https://scholarlycommons.law.case.edu/cgi/viewcontent.cgi?params=/context/jolti/article/1172/&path_info=auto_convert.pdf > Accessed on 10^th September, 2025). China pursues a state-driven, techno-industrial strategy linking AI to national development goals(Kyle Chan, Gregory Smith, Jimmy Goodrich, Gerard Dipippo, Konstantin F, Pilz ‘China’s Evolving Industrial Policy for AI’ (Rand.org, 26^th June, 2025) < www.rand.org/pubs/perspectives/PEA4012-1.html > Accessed on 10^th September, 2025). The EU relies on its regulatory power (“the Brussels effect”) to set global supplier standards (Marco Almada, Anca Radu, ‘The Brussels Side-Effect: How the AI Act Can Reduce the Global Reach of EU Policy’ (Cambridge.org, 19^th February, 2024) < www.cambridge.org/core/journals/german-law-journal/article/brussels-sideeffect-how-the-ai-act-can-reduce-the-global-reach-of-eu-policy/032C72AEC537EBB6AE96C0FD90387E3E > Accessed on 10^th September, 2025). Together, these approaches create a patchwork of norms that countries and companies must navigate. (To be continued).

THOUGHT FOR THE WEEK

“Like all technologies before it, artificial intelligence will reflect the values of its creators. So inclusivity matters – from who designs it to who sits on the company boards and which ethical perspectives are included”. (Kate Crawford).