May 2026 · Observation (I)
Izhar Ahmad
INITIAL TAKE
Historically, digital infrastructure environments could absorb isolated weaknesses such as expensive power, inefficient cooling, limited fiber density, or permitting delays because these pressures did not always intensify simultaneously. AI infrastructure changes this equation. Global cloud and AI workloads place concurrent pressure on energy systems, cooling environments, interconnectivity, land access, and timelines.
As AI workloads scale even further, existing regional infrastructure differences are becoming more important in determining where AI capacity can realistically grow. Europe is shaped by grid and permitting bottlenecks. The Middle East and North Africa (MENA) region faces energy, water and thermal constraints. Southeast Asia must manage heat, humidity, and climate-related operational stress, while in the rest of Africa longer-term opportunity depends heavily on infrastructure coordination and deployment readiness, a key first component of which is policymaking.
AI infrastructure is less forgiving because operational pressure accumulates across multiple infrastructure layers at the same time. The emerging divide is increasingly determined by regional infrastructure tolerance thresholds. I refer to these regional infrastructure tolerance thresholds as the level of AI infrastructure expansion that power systems, cooling environments, interconnectivity networks, land access conditions, and regulatory frameworks can absorb without creating additional operational or deployment complications.
Apart from AI development conditions, AI infrastructure launch timelines are also important. Infrastructure scale alone is insufficient if power delivery, permitting, cooling systems, interconnection capacity, and operational readiness cannot move forward within commercially viable timeframes.
What we are observing, and on which multi-stakeholder alignment is merited, is that AI infrastructure is increasingly separating digital geography from economic geography. The next major AI infrastructure hubs may increasingly emerge outside traditional digital and economic centers, not because demand has shifted, but because infrastructure pressure has.
Yours in alignment,
Izhar Ahmad
WHERE INTELLIGENCE & ALIGNMENT EMERGES
Across Africa, Europe, MENA, and Southeast Asia, a different infrastructure logic is becoming visible: AI infrastructure growth is increasingly determined by how effectively regions can align infrastructure conditions across land access, power, cooling, connectivity, and deployment readiness supported by the right policy and regulatory frameworks.
On their own, individual infrastructure strengths are no longer sufficient. That is, energy availability without thermal sustainability, connectivity without grid expansion, land availability without deployment readiness, or investment without regulatory incentivization, for example, all increasingly create limits on scalable AI capacity. Infrastructure pressure is just not isolated. Power systems, cooling environments, interconnectivity networks, deployment timelines, and regulatory processes increasingly influence one another simultaneously. Pressure accumulation in one layer increasingly affects infrastructure performance and scalability in another.
Regions capable of aligning energy delivery, cooling feasibility, permitting, interconnection capacity, and deployment execution more efficiently may increasingly gain advantages in attracting long-term AI infrastructure investment, because capacity will move toward regions where infrastructure tolerance thresholds are higher and where infrastructure conditions can remain manageable under sustained AI demand.
Traditional regional digital hubs continue to hold structural advantages in connectivity, enterprise concentration, and ecosystem maturity. However, at the same time, infrastructure pressure within these environments is increasing the importance of secondary and emerging markets capable of absorbing additional AI capacity more efficiently. This is how a new digital geography is taking shape before our eyes. We can expect to see such geographical shifts emerge in Africa in the near term.
FINAL TAKE
The next phase of AI infrastructure growth may depend less on isolated infrastructure advantages and more on how effectively regions manage simultaneous infrastructure pressures across multiple dimensions. With interest now rising in sovereign AI development across the world, as is being observed from West Africa to Central Asia to South Asia, the new digital geography will likely favor regions capable of timely and intelligently aligning across all infrastructure tolerance thresholds.
As AI infrastructure scales, infrastructure tolerance thresholds will determine where AI investment will actually take place and which cities, regions, or markets are capable of building AI capacity. Every market needs this capacity, but not all are ready yet to build and sustain it.
Intelligence accumulates as perspectives diverge.
Alignment emerges as conditions converge.
References
¹ International Energy Agency (IEA), Energy and AI — https://www.iea.org/reports/energy-and-ai
² International Energy Agency (IEA), Energy Demand from AI — https://www.iea.org/reports/energy-and-ai/energy-demand-from-ai
³ CBRE, Global Data Center Trends 2025 — https://www.cbre.com/insights/reports/global-data-center-trends-2025
⁴ Reuters, Power supply constraints slowing EMEA data centre rollout — https://www.reuters.com/business/energy/power-supply-constraints-slowing-emea-data-centre-rollout-report-says-2025-11-06/