The Not-So-Public Digital Infrastructure That Runs Your AI
- Oct 14, 2025
- 1 min read
👉 There are 1.48 million km of submarine cables (30X the circumference of the Earth) connecting countries to move DATA
👉 Example 99% of data to the outside world in UK is carried through 60 cables
👉 95% of data still move by cables (sea or land)
👉 Most of this infrastructure is owned by the private sector (99%)
👉 A disruption of a cable leads to slow internet - take 2008, 2 Mediterranean cable were cut affecting 14 countries (100% of Maldives and 82% of India)
👉 Few players in fabrication: USA, France, China, Japan...
👉 Choke points - Think Egypt - 90% of Eurasian traffic flows through Egypt (via Telecom Egypt)
👉 Data servers store data and allow for cloud compute - 50% of which are hypescalers (more than 1 million Sq. Feet)
👉 Only 25 companies qualify as hyper scalers - includes your BigTech
👉 Hyperscalers are now looking at Energy infrastructure to fuel these monsters
👉 Cables lifecycle is around 25 years, Hyperscalers servers need to be replaced every 3-8 years -- E-waste is the fastest growing waste - we do not have any AI National Strategies that outline e-waste recycling!!! Very few tech companies are investing directly in e-waste recycling!!!
👉 E-Waste affects health, contaminates water, food chain, is a result of dirty mining... and also is trafficked to poorer countries by richer ones!
The article on the not-so-public digital infrastructure that runs AI raises important questions about transparency and accountability in technology systems. I appreciate the critical lens on hidden infrastructures. An artificial intelligence (ai) course & workshop for managers in Cairo, Egypt could help IT professionals understand the foundational systems behind AI, assess risks, and design more transparent and ethical AI solutions.
The digital infrastructure powering AI operates largely out of sight — data centres, undersea cables, cloud computing networks, and energy grids that most managers never think about until they fail, exposing just how fragile the foundations of AI-dependent organizations truly are. For managers building genuine AI literacy beyond the surface level, the artificial intelligence (AI) seminar & course for managers in Cairo, Egypt offers the technical frameworks and strategic thinking needed to understand, manage, and build resilience around the infrastructure that keeps AI systems running with genuine operational depth.