Tech Dreams, Power Nightmares

AI Dreams Need Electricity

Nigeria’s push to become a digital powerhouse gained global attention on February 12, 2025, when President Bola Tinubu announced a partnership discussion with Google to boost cloud infrastructure, Artificial Intelligence (AI) research, and workforce development. However, the announcement quickly sparked debate, with many Nigerians questioning how the country can lead in technological advancements while persistent power shortages leave even hospitals in darkness for months.

Nowhere is this contradiction more evident than in Lagos, Nigeria’s bustling tech capital, where AI startups are emerging rapidly, promising innovations in fintech, healthcare, and education. Yet, on any given day, a power outage can bring operations to a standstill, forcing businesses to rely on costly generators. Despite the nation’s ambition to become an AI powerhouse, its fragile electricity grid remains a major obstacle.

The government, backed by private investment, is positioning the country as Africa's leading AI and tech hub. But this vision is faced with a harsh reality: an unreliable power supply that stifles progress. Advanced technology and digital innovation depend on stable electricity and reliable infrastructure—two essentials that Nigeria struggles to provide.

Frequent blackouts not only drive up operational costs but also limit technological growth. The disparity raises important considerations about resource allocation. Should a country facing continuous electricity shortages prioritize advanced technology initiatives before solving basic infrastructure needs?

This ambition to lead in artificial intelligence continues to gain traction across Nigeria’s tech ecosystem, where new startups are building solutions for finance, health, and agriculture. Innovation hubs, including CcHub, have become central to this push, offering space, funding, and mentorship to early-stage ventures. At the same time, the government’s National Digital Economy Policy and Strategy (NDEPS) seeks to create an environment where local talent can develop AI-driven solutions tailored to Nigeria’s unique needs.

International investors have taken notice, with increasing interest in startups tackling local and regional problems through machine learning and automation. Proposed AI regulations are also taking shape, signaling an effort to create clearer rules for data collection, privacy, and ethical development in a rapidly expanding sector. While policies and investments point to optimism, the infrastructure powering these innovations has not kept pace with the sector’s growth.

Nigeria’s electricity supply struggles to support even basic digital operations, let alone the demands of artificial intelligence. Despite an installed generation capacity exceeding 12,000 megawatts, only around 4,000 megawatts consistently reach users across the country. This limited output serves a population of over 200 million, placing immense strain on businesses and households alike. For AI startups relying on continuous access to power for data processing and cloud-based operations, unreliable supply presents a serious obstacle.

Self-generated electricity, typically through diesel generators, has become a lifeline for many technology firms. The cost of running these generators is substantial, cutting into budgets that would otherwise fund product development, hiring, or research. AI startups, particularly those still establishing themselves, often struggle to absorb these costs while maintaining competitive pricing and sustainable growth. Continuous computing, essential for model training and data analysis, becomes a financial burden that larger or better-funded companies may withstand—but smaller firms often cannot.

Between January and December 2024, the national grid collapsed 12 times, with each failure disrupting businesses and research efforts across the country. In a single month, technology firms reported close to 30 power cuts, each lasting an average of three hours. Every interruption slows development cycles, disrupts collaboration, and adds further pressure to already strained budgets. Some companies with the means to do so rely heavily on alternative energy, but many smaller players either reduce operations or shut down entirely during prolonged outages.

Costs Beyond Money

This dependency on constant electricity creates a difficult balance for AI firms already managing high data processing demands. Training large-scale models alone requires a level of power consumption that few Nigerian businesses can sustain without external support. Training OpenAI’s GPT-3, for instance, consumed around 1,287 megawatt-hours of electricity — roughly 100 times the annual usage of a typical American household. While global firms like Google and Microsoft running similar projects can lean on reliable grids and cleaner energy sources, Nigerian firms face a different reality where power supply interruptions are frequent and renewable infrastructure remains underdeveloped.

With limited access to stable electricity, AI companies turn to diesel and petrol generators to keep servers running and data flowing. For small and mid-sized firms, energy already accounts for an estimated 36% of total operating expenses — a figure far higher than the less than 10% spent by firms in advanced economies. The high cost of self-generated power forces businesses to divert funds that could otherwise support product development, talent acquisition, or market expansion. Startups relying on continuous machine learning training cycles face even greater financial strain, as the energy demands of these processes surpass those of traditional software development.

Total spending on fuel and generator maintenance reaches billions of naira annually across the broader business sector, placing energy among the largest operational costs for technology firms. This creates an uneven playing field, where local firms compete against global players benefiting from cheaper, cleaner, and more reliable electricity. The financial pressure ultimately reduces the ability of Nigerian AI startups to scale or attract sustained investment, especially when potential backers weigh these infrastructure costs against opportunities in other regions with more favorable conditions.

The environmental consequences of generator dependence add further complexity. In urban areas like Lagos, widespread generator use contributes to rising levels of air pollution, increasing the incidence of respiratory illnesses and placing further strain on public health systems. Fossil fuel reliance also amplifies greenhouse gas emissions, with 79.5% of Nigeria’s electricity coming from non-renewable sources. As a result, the same AI sector expected to drive digital advancement contributes directly to environmental degradation, creating a contradiction between technological ambition and environmental responsibility.

Energy policy gaps and outdated distribution infrastructure leave businesses with few options beyond self-generation. Without major investment in renewable energy and modern grid systems, AI’s expansion risks compounding environmental and public health pressures, adding another layer of complexity to Nigeria’s push to compete in the global digital economy.

This growing interest in renewable energy highlights the pressing need for reliable alternatives, especially for tech companies running data-intensive systems. Nigeria receives an average of 5.5 kilowatt-hours per square meter of sunlight daily, offering ample potential for solar power to bridge the supply gap. Some technology hubs and early adopters already depend on solar installations to keep systems running when grid electricity fails. Despite this advantage, the total installed solar capacity stood at only 385.7 megawatts peak in 2024, falling far below the level needed to support broader industrial and technology-driven power demands.

Some companies, including Arnergy, Husk, and Lumos, supply solar systems tailored for businesses looking to reduce fuel consumption from diesel generators. Data centers like Daystar Power and Afrinet Power have started incorporating solar into their energy mix, aiming to secure consistent electricity for continuous computing tasks required in advanced data processing. Off-grid solutions, including solar mini-grids and self-sufficient power systems, have already brought electricity to several rural areas, helping small businesses operate independently of the fragile national grid.

Solar-based projects offer a promising way to lower operating expenses and reduce reliance on high-emission fuels. However, the financial burden of deploying large-scale renewable systems remains prohibitive for many startups and mid-sized firms. Solar panels, inverters, battery storage, and installation costs often require substantial upfront investment, which few early-stage companies can afford without external financing.

Countries like Kenya provide useful comparisons, where the country’s investment in renewable energy, especially geothermal and solar, has enabled a more stable power supply, supporting the growing tech ecosystem. Renewable-friendly policies and financial incentives have significantly accelerated adoption. Nairobi’s technology sector benefits from tax breaks and subsidized loans for clean energy projects, making it easier for companies to transition away from expensive backup power. In Nigeria, however, the high upfront cost of solar infrastructure makes it difficult for startups and businesses to adopt large-scale renewable energy solutions. Government incentives remain limited, with existing policies largely favoring conventional energy sources over solar and other renewables.

Weak grid infrastructure also complicates renewable integration, especially outside major urban centers. Areas without modern transmission and distribution networks cannot often connect new solar projects or efficiently distribute surplus power. This infrastructure gap limits the potential benefits of distributed solar systems, even where technology firms are willing to invest.

A more effective public-private partnership framework could reduce financial risks for renewable projects, especially those targeting technology clusters and industrial zones. Clearer regulatory support and improved access to affordable financing would give businesses the confidence to adopt solar and other clean energy solutions, allowing AI and data-driven enterprises to operate without constant dependence on costly, polluting generators.

Plans vs Practical Implementation

This regulatory landscape is shaped heavily by the Nigerian Electricity Regulatory Commission (NERC), which oversees licensing, tariffs, and market rules following the country’s power sector privatization. While the Electricity Act of 2023 introduced provisions allowing states to generate and distribute power independently, practical implementation remains uneven. Several states lack the technical capacity and financing required to develop their energy projects, slowing efforts to decentralize the electricity supply.

Despite roadmaps like the Power Sector Recovery Program (PSRP), intended to enhance grid performance, electrification rates remain low, with only 60% of the population connected to the national grid. Infrastructure limitations and regulatory inconsistencies continue to discourage sustained private sector investment, particularly in renewable energy. For technology companies relying on uninterrupted power, these conditions make long-term planning difficult, with businesses forced to hedge against grid failures through costly alternative energy solutions.

As Nigeria positions itself as a technology hub, the National Digital Economy Policy and Strategy (NDEPS) outlines ambitious goals for innovation and investment attraction. Proposed AI-specific policies aim to create a favorable environment for machine learning research and deployment. However, reliable electricity remains conspicuously absent from these strategies, leaving a foundational gap in the broader digital transformation agenda. Over 92 million people still lack stable power access, meaning even the most promising AI policy framework risks becoming irrelevant if essential infrastructure needs go unaddressed.

Energy-intensive computing requires far more reliability than the existing power network can provide. With constant blackouts and soaring fuel costs for backup power, technology companies face severe constraints, limiting how much they can invest in advanced data processing and innovation. Without reliable access to electricity, data centers, research labs, and software development firms operate under unsustainable conditions, pushing smaller businesses to either downscale operations or relocate to regions with better infrastructure.

Experiences from Rwanda and South Africa demonstrate how intentional policy alignment between energy and technology development can help build stable digital ecosystems. Rwanda’s investments in hydropower and off-grid solar contributed to a 70% electrification rate, enabling consistent electricity access in Kigali’s innovation hubs. South Africa’s Renewable Energy Independent Power Producer Procurement Program (REIPPPP) allowed private firms to supply electricity directly to the grid, reducing pressure on Eskom while opening new financing channels for renewable projects. Both approaches underscore how deliberate energy sector reform, combined with targeted technology policies, creates a more resilient environment for digital industries. Without addressing these core structural gaps, any effort to scale AI or digital innovation risks becoming largely theoretical.

This ongoing dependence on outdated power infrastructure continues to limit how effectively AI-focused businesses can operate. Unreliable electricity forces technology firms to either scale down computationally intensive projects or relocate to regions with more stable grids, leaving local innovation clusters struggling to compete with global counterparts. While some firms in urban centers manage to offset grid failures with alternative energy, this option remains financially out of reach for many early-stage startups, especially those outside Lagos and Abuja.

Grid modernization remains a critical yet largely neglected component of Nigeria’s broader digital transformation strategy. Aging transmission lines, poor maintenance practices, and widespread energy theft combine to create an unstable supply environment that undercuts technological advancement. Even when electricity reaches businesses, voltage fluctuations often damage sensitive hardware, adding to operational expenses for firms already grappling with high fuel costs. Currently, Nigeria ranks 103rd on the Global AI Adoption Readiness Index, highlighting its struggle to integrate AI due to foundational infrastructure issues.

Diversifying Nigeria’s energy mix could ease some of these pressures, but renewable energy uptake within the technology sector remains limited. Solar and wind projects often face lengthy approval processes, inconsistent regulatory support, and high initial investment requirements. Without policy changes that prioritize clean energy adoption for tech-dependent businesses, AI-driven companies will continue relying on polluting, expensive fossil fuel generators to meet their basic power needs.

Public and private sector partnerships could help close these gaps, but existing collaboration mechanisms often lack clear accountability structures or enforceable targets. Current incentives for private investment in renewable energy are fragmented across multiple agencies, creating uncertainty for firms interested in building sustainable power infrastructure. Without simplifying these processes, renewable energy adoption among AI firms will likely remain sporadic and limited to companies with the resources to navigate Nigeria’s complex regulatory environment.

Existing government-led AI initiatives, including those housed at the National Centre for AI and Robotics, face similar limitations. Research projects requiring continuous data processing, model training, or cloud-based collaboration remain vulnerable to sudden power interruptions, disrupting research timelines and limiting Nigeria’s ability to produce homegrown AI solutions that address local challenges. Without stable electricity, these initiatives risk becoming symbolic rather than functional drivers of technological progress.

Countries with comparable development goals, such as Rwanda, have demonstrated how linking renewable energy expansion directly to digital innovation policies can help build resilience into national technology ecosystems. Until Nigeria adopts a similarly coordinated approach, AI applications will remain concentrated within select industries and locations, further widening the digital divide between well-funded technology hubs and regions where infrastructure gaps continue to block meaningful participation in the broader digital economy.

Achieving any meaningful progress in artificial intelligence or broader technological innovation depends entirely on addressing the underlying power crisis that continues to plague Nigeria’s economy. Businesses can no longer afford to navigate a system where electricity remains both unreliable and prohibitively expensive, especially when high-performance computing depends on stable, continuous power.

For startups, research centers, and technology hubs hoping to harness AI for real-world applications, the absence of dependable electricity limits their ability to attract investment, scale operations, and retain local talent. As long as energy insecurity persists, the cost of innovation will remain artificially high, locking many promising ideas out of the market before they have a chance to succeed.

Prioritizing sustainable power solutions, simplifying regulatory frameworks, and aligning energy policies with digital transformation goals offer Nigeria a realistic path forward. Without treating reliable electricity as the foundation for technological progress, AI will likely remain a tool for only the wealthiest firms, leaving most businesses and communities unable to fully participate in the country’s digital future.