February 9, 2026
In an interview published by Gulf Business, Mohamed Al-Masry, vice president of Middle East and Africa at Hitachi Energy, said artificial intelligence (AI) is rapidly becoming an energy and infrastructure challenge, reshaping the way power grids are designed, financed and operated. As hyperscalers deploy large-scale AI data centers, power demand has skyrocketed from tens of megawatts to hundreds of megawatts, approaching gigawatt scale, and with completely different load patterns than traditional industrial users. It is now becoming clear that grid capacity, permit schedules, and equipment bottlenecks are real brakes on AI growth in many mature markets.
Unstable AI loads challenge grid stability
AI workloads combine extremely high power demands with rapid and erratic load behavior, putting new pressure on the grid. While traditional data centers typically operate at 1 to 10 megawatts with steady demand, AI data centers can range in power from 100 to 500 megawatts, with some approaching gigawatts. The main challenge is the rapidly fluctuating demand from AI training, which requires repeated cycles of feedback and refinement. These fast fluctuations make it difficult to manage frequency, voltage, and power quality, leading some operators to classify AI data centers as noncompliant loads.
Capacity bottlenecks and connection delays
The most severe bottlenecks are in the regions where AI data centers cluster the fastest. In Europe, power grid capacity is approaching capacity in hub cities such as Amsterdam, Frankfurt and Dublin. In parts of Asia such as Tokyo and Mumbai, developers have been securing power for years. Similar clustering is causing some regions of North America to run out of capacity, while China is shifting its AI infrastructure westward. The International Energy Agency (IEA) estimates that up to 20% of data center capacity could experience connectivity delays between 2025 and 2030. AI projects move quickly, but grid permitting can take five to 10 years, and equipment such as transformers currently takes two to four years to deliver.
The Middle East stands as a competitor in AI infrastructure
The Middle East, and the GCC in particular, is positioning itself as a contender for next-generation AI infrastructure due to its energy scale, capital, and faster execution. The region’s advantage comes from the combination of large-scale energy availability and rapid integrated infrastructure provision. Proximity to generation is important because existing grids were not designed with the scale and variability of AI in mind. Sovereign wealth funds can directly support multi-gigawatt computing, grid and power generation projects, avoiding delays seen elsewhere. AI is a national priority across the Gulf, enabling faster permitting and long-term planning.
Grid technology and planning needs to evolve
The grid can no longer treat AI data centers as a steady load. AI training will cause rapid demand fluctuations on an industrial scale, requiring more power electronics, digital controls, and power quality technologies to maintain stability. This has increased the demand for high voltage switchgear, disconnectors and generator circuit breakers. While transmission planning must move to a cluster-based approach, on-site and on-grid battery energy storage is essential to smooth rapid load changes. Technologies such as STATCOMs, voltage regulators, and harmonic filters are becoming increasingly important.
Saudi Arabia and UAE National Strategy
Saudi Arabia treats AI as critical infrastructure under Vision 2030, linking data centers to renewable energy and grid expansion. The UAE is closely integrating its AI strategy with its energy policy. This alignment between AI strategy, energy planning, and grid development makes both connectivity risk and regulatory uncertainty clearer and significantly reduced for investors.
Requirements for successful AI implementation
The success of an AI site first depends on preparing a large-scale, real-world grid. Hyperscalers require solid long-term capacity commitments and clear grid reinforcement plans. The Gulfs advantage will only be maintained if expansion is planned early and aligned with AI demands. On-site solar power generation and energy storage improves resiliency. Regulatory certainty is key and for sites located in established digital zones or free zones, streamlined permitting reduces execution risk. Long-term planning should consider climate, scalability, and connectivity to submarine cables and Internet exchanges.
Source: IndexBox Market Intelligence Platform
