Home Technology Data Centre Insights: AI data centres and the Middle East advantage The Middle East is uniquely positioned to become a global AI DC powerhouse, combining strategic geography, low costs, policy momentum, and geopolitical alignment by Thibault Werlé August 22, 2025 Follow us Follow on Google News Follow on Facebook Follow on Instagram Follow on X Follow on LinkedIn Image: Getty Images/ For illustrative purposes AI data centres (AI DCs) are the essential ‘hardware backbone’ of the global AI race, with the Middle East emerging as a key player with the region’s most recent initiative launches on the global map. The demand for AI computing power has grown exponentially. Despite disruptors like Deepseek and improvements in chip efficiency, industry experts predict that increasing AI complexity and adoption will continue to fuel its steep upward trajectory. In our recent publication ‘Breaking Barriers to Data Centre Growth’, we expect that global data centre power demand will reach 127GW by 2028, up from 60GW in 2023. Nearly 35 per cent of the total DC power demand in 2028 will support GenAI. More than 90 per cent of this will be used for inferencing – which makes sense, as inferencing encompasses the repeated, widespread use of a given AI model. Skin in the game, chips on the table: The Middle East AI push is on The Middle East is quickly emerging as a key AI DC nexus. Major governments in the region are pursuing large-scale national AI strategies and investments. New developments include: In Saudi Arabia Launch of PIF-backed HUMAIN, KSA’s AI champion that will invest and operate across the AI value chain HUMAIN x NVIDIA’s partnership to build up to 500 MW AI DC over the next five years $10bn HUMAIN x AMD collaboration for 500 MW AI DC over the next five years $5bn HUMAIN x AWS partnership for an advanced AI zone $5bn DataVolt establishment of the region’s first truly sustainable, net-zero AI campus in Oxagon $1.5bn Groq x Aramco partnership to build the world’s largest AI computing centre In the UAE 5GW-power AI campus in Abu Dhabi as part of the US-UAE AI Acceleration Partnership, built by G42 and American companies. The DC will cover 10 square miles and will be the largest AI DC outside of the US 500,000 GPUs to be imported, 20 per cent for G42 and the rest for US companies building AI DCs 160MW data centres across Ajman and Abu Dhabi by Khazna Data Centres The Middle East is emerging as a crucial hub for AI data centre investment and innovation, offering several compelling advantages. The region’s strategic location allows it to serve around three billion people (~40 per cent of the world’s population) within a 2,000-mile radius, covering Europe, Asia, and Africa. This geographical advantage is ideal for hosting non-latency-sensitive inferencing services at scale and positions the Middle East as a significant provider of AI computing services to the Global South. Cost efficiency is another substantial benefit, with leasing costs up to 90 per cent lower than global averages and some of the lowest power tariffs worldwide. Additionally, emerging data centres in the region are showcasing state-of-the-art innovation, leveraging advanced cooling technologies targeting leadership in power usage efficiency (PUE) rankings and drastically reducing total cost of ownership (TCO). The region also offers quicker time-to-market benefits, thanks to countries like Saudi Arabia and the UAE, which have dedicated investment teams, streamlined regulatory processes, fast-track permitting, and special economic zones that simplify licensing and shorten DC completion times. As for infrastructure, the GCC boasts expansive land banks, scalable power supplies, and planned connectivity expansions with the Fibre in Gulf (FIG) submarine cable project, expecting capacities of up to ~720 Tbps. Winning takes more than power Countries around the world are racing to capture the growing AI infrastructure demand, recognising both its economic and strategic benefits. To establish the Middle East as a leader in the AI competition, governments and businesses need to focus on four critical areas. First, lowering the total cost of ownership (TCO); as it is the primary value metric for international investors, with ongoing competitiveness and cost reductions central to a country’s value proposition. Second, attracting a diverse mix of tenants, such as GPU-as-a-Service providers, hyperscalers, and colocation providers, is crucial to creating a thriving ecosystem that mitigates demand/supply risks and maximises utilisation. Third, securing ecosystem partnerships is necessary, with collaboration among chipset companies, component providers, and research institutions strengthening local value propositions and the AI data centre value chain. Finally, developing and retaining talent is essential, as a skilled workforce is critical for driving innovation in AI models, platforms, and applications, ensuring long-term leadership beyond the infrastructure itself. The bottom line As AI reshapes economies and geopolitics, infrastructure is the frontline. The Middle East is uniquely positioned to become a global AI DC powerhouse, combining strategic geography, low costs, policy momentum, and geopolitical alignment. But seizing this moment will require more than megawatts, it will demand partnerships, ecosystem depth, and a long-term talent pipeline The writer is MD and partner, Boston Consulting Group (BCG) & Marc Nasr, and MD and partner, BCG. Read: Taking AI-driven data centres into the future Tags AI BCG Data centres Insights Technology