The AI Digital Divide Is Widening. Most Coverage Isn't Tracking It.
Author: Protik Ganguly
The AI digital divide is widening — and most coverage isn't tracking it. The promise was elegant: AI democratises opportunity. A student in Lagos gets the same research tools as a student at MIT. A farmer in Bihar gets the same market intelligence as a trader in Mumbai. The technology is neutral. It goes where the internet goes.
The internet goes to 38% of Africa. Mobile data in Sub-Saharan Africa costs fourteen times more than in Europe (IEEE, 2026). Developing nations hold 0% of global supercomputer capacity (World Bank, 2025). Africa accounts for less than 1% of global data centre capacity despite being home to 18% of the world's population (CSIS, 2025). The promise of AI as leveler runs directly into the physics of infrastructure.
The divide is not only between nations. It runs through wealthy ones too. In the United States, 19 million Americans still lack broadband internet — disproportionately concentrated in rural states like Louisiana, North Carolina, West Virginia, and Mississippi (FCC, 2024). Connectivity gaps persist across rural Europe too — eastern Germany, southern Italy, Greece, and Romania. AI tools requiring reliable internet, cloud access, and subscription fees are not equally available in rural Louisiana and San Francisco — even within the same country and nominally the same opportunity landscape.
The numbers on global benefit distribution are starker. If current trends continue, only 3% of AI's projected $19.9 trillion in global economic benefits will go to Latin America. A mere 8% will go to the combined populations of Africa, Oceania, and other Asian markets excluding China (CSIS, 2025). The technology is not neutral. It flows where the infrastructure already exists.
The geopolitical dimension is accelerating this. DeepSeek usage in Africa is estimated at 2 to 4 times higher than in other regions — because DeepSeek removed the cost barriers Western models maintained (Microsoft Research, 2026). Huawei is actively deploying AI infrastructure across African markets. Microsoft has invested $8 billion in Global South data centres. Both are welcome. Both are someone else's infrastructure, on someone else's terms.
This is not inevitable. India's BHASHINI platform demonstrates what deliberate intervention looks like — 100 million AI inferences monthly across 22 Indian languages, built by government because no private company had commercial incentive. India Stack — now processing more transactions than Visa and Mastercard combined in India — is the template.
AI will level the playing field eventually. The question is who builds the level. The countries that build their own will shape what leveling means. The countries that simply receive it will inherit a playing field designed for someone else's game.
References
FCC. (2024). Broadband deployment report. https://www.fcc.gov/reports-research/reports/broadband-progress-reports
IEEE / arxiv. (2026). Bridging the AI divide in Sub-Saharan Africa. https://arxiv.org/pdf/2601.06145
Microsoft Research. (2026, January). Global AI adoption in 2025: A widening digital divide. https://www.microsoft.com/en-us/research/wp-content/uploads/2026/01/Microsoft-AI-Diffusion-Report-2025-H2.pdf
World Bank. (2025, December). World Bank warns of widening compute divide as poor nations hold 0% of global supercomputer capacity. https://medium.com/@impactnews-wire/world-bank-warns-of-widening-compute-divide-as-poor-nations-hold-0-of-global-supercomputer-dca21b7c6069
Related Articles
Embrace the Uncomfortable — The Case for Walking the Walk
• The gap in productivity, income, and relevance between early adopters and late adopters of transformative technologies tends to widen faster than expected and close slower than hoped. • Companies that built strategies around early versions of the internet, smartphones, and cloud computing captured disproportionate value, while those who waited found the territory already claimed. • Early adopters of AI tools are reportedly experiencing productivity gains of 20 to 40% on drafting, research, and analysis tasks, with companies that redesigned workflows before deployment showing significant measurable benefits.
The Uncomfortable Middle — What the Transition Actually Costs
• The transition to the Industrial Revolution was marked by the Luddite uprisings, where textile workers destroyed machinery due to desperation about their skills being made obsolete. This event highlights the gap between disappearing jobs and new, yet-to-be-invented ones. • The current AI transition is expected to replace many jobs, with computer scientist Geoffrey Hinton predicting significant capabilities in 2026. AI systems are rapidly completing tasks that previously took longer, with a recent MIT analysis finding they can handle nearly 12% of tasks tied to the US labor market. • Entry-level white-collar roles are being affected first by AI, with tasks such as drafting, research, and analysis being handled by machines, potentially hollowing out junior positions that once performed these tasks.
AI Is Making Tasks Faster. It Isn't Making Companies Richer.
• American companies are expected to spend $675 billion on AI infrastructure in 2026, a 63% increase from last year, but the financial returns on this investment are unclear. • Studies have shown that AI can make tasks faster and improve output quality, but this does not necessarily translate to increased company-level productivity. • Only a small percentage of companies, such as 21% of S&P 500 companies, can cite a measurable AI benefit, and many executives struggle to confidently measure AI return on investment.