Every generation of technology arrives with a familiar promise: that it will change everything.
Artificial intelligence is the latest to carry that burden. It is described as transformative, inevitable, and in some cases uncontrollable. Capital is flowing at extraordinary speed, companies are being valued on future dominance, and governments are racing to respond. Alongside this momentum, however, a more cautious conversation is emerging — about uncertain returns, inflated expectations, and the growing carbon footprint of the very systems being scaled.
This moment is not new. It is part of a recurring cycle.
Across decades, technologies have been introduced with predictions of total disruption. Digital media was expected to eliminate print. E-commerce was meant to replace physical retail. Television was supposed to end radio, just as streaming would later displace traditional broadcasting. More recently, NFTs and the metaverse were positioned as foundational shifts in how we assign value and interact.
Yet in each case, reality proved more complex.
Print persists. Retail evolved into hybrid models. Broadcast and streaming now coexist. Even remote work, once declared the future of all work, has settled into a negotiated balance with in-office presence. Technologies do not simply replace systems; they are absorbed into them — reshaping, redistributing, and often falling short of their most ambitious claims.
Even where the promise is real, execution is uneven. The dot-com boom transformed the global economy—but only after a painful correction. Cloud computing became indispensable — but not as a total replacement for existing systems. Carbon trading, conceived as an elegant market solution to climate change, has delivered mixed results at best, constrained by fragmented implementation and inconsistent incentives.
These are not failures of innovation. They are failures of expectation.
Artificial intelligence now sits at the intersection of genuine capability and familiar overreach.
There is little doubt about its potential. AI can process information at scale, augment decision-making, and unlock efficiencies across industries. But it also raises difficult questions—about energy consumption, environmental sustainability, governance, and equitable access. Its carbon footprint introduces a new layer of complexity for a world already struggling to align growth with climate responsibility.
For Africa, the stakes are more immediate.
This is not simply about participating in a global technology cycle. It is about whether AI can be applied in ways that accelerate development in sectors where progress has historically been constrained.
The opportunity is clear.
In public health, AI can strengthen disease surveillance, improve diagnostics, and enhance pandemic preparedness—capabilities that remain uneven across the continent. In education, it can help scale access to quality learning despite shortages in teaching capacity. In energy, it can optimise grid performance, reduce losses, and support the integration of renewable systems in fragile networks. In agriculture, it can improve yield forecasting, climate adaptation, and supply chain efficiency for millions of smallholder farmers.
These are not speculative use cases. They are aligned with structural needs.
But history suggests that potential alone is not enough.
The risk is not that AI will fail, but that it will be misapplied—deployed too broadly where it adds little value, and too cautiously where it could have a meaningful impact. Without clear strategy, it could follow the path of previous technological waves: attracting attention and capital but delivering uneven and fragmented outcomes.
Avoiding that outcome requires discipline.
For policymakers, especially in Africa where the impact of policy errors can be existential, this means focusing on targeted deployment rather than symbolic adoption. AI strategies should be anchored in specific sectors with measurable outcomes, not broad declarations of intent. Regulatory frameworks—particularly around data governance, privacy, and cross-border flows—must evolve alongside deployment, not after the fact.
For investors and business leaders, it means resisting the pull of generalised narratives. The most durable opportunities will not be in AI as a theme, but in AI applied to defined problems within existing constraints. Capital should flow to solutions that are grounded in economic reality, not technological enthusiasm.
For both, infrastructure cannot be an afterthought. AI depends on energy, connectivity, and compute capacity—areas where gaps remain significant. The conversation about AI’s carbon footprint is especially relevant in African contexts, where energy systems are either inadequate or already under strain. Scaling intelligence without addressing power is a contradiction that cannot hold.
Finally, coordination is key. Governments, private sector actors, and development institutions must align around the critical, practical and optional use cases rather than duplicated, fragmented efforts. Without this prioritisation and synergy, even the most promising technologies risk underperformance. And for Africa failure with AI amplifies existing disadvantages and compounds poverty and angst.
The carcasses of past technologies offer a useful perspective. They remind us that disruption is rarely absolute, that markets correct, and that systems adapt more slowly than narratives suggest. They also show that development is hardly linear, but could be vaulted forward with technology, layered over reality.
Artificial intelligence will shape the future. But it will not do so on its own terms. Its impact will depend on the quality of the choices made now—where it is applied, how it is governed, and whether it is aligned with real economic and social priorities.
For Africa, the question is not whether to adopt AI, but how to do so with clarity and intent. Because in the end, what endures is not what promises the most—but what works.
- business a.m. commits to publishing a diversity of views, opinions and comments. It, therefore, welcomes your reaction to this and any of our articles via email: comment@businessamlive.com
Southeast: Between building PHCs and delivering healthcare