Commentary by Mona Yacoubian and Samuel Zabin

Published February 27, 2026

With conflict looming in the Gulf, the region's nascent AI infrastructure could be imperiled, further raising the stakes of a regional war. The prologue to the Pax Silica Declaration, a Department of State initiative on AI and supply chain security, states, "If the 20th century ran on oil and steel, the 21st century runs on compute and the minerals that feed it." Qatar and the United Arab Emirates (UAE) are among the declaration's 10 signatories, highlighting the Gulf's increasing relevance as a locus of this strategic technology. If compute-the infrastructure and resources that power AI-is indeed the new oil, the Gulf's growing AI role could reshape U.S.-Gulf ties, introducing new benefits, risks, and interdependencies.

The parallels between oil and compute are notable. Both require massive upfront investments that demand sustained engagement among firms and partner nations. Both resources are subject to export controls, regulatory pressure, and intentional disruption. However, unlike oil, compute doesn't have a single geographic capture point; instead, it relies on several strategic chokepoints, such as chips, software, energy, and data centers. The absence of a single capture point inhibits historic embargo-style dominance, yet the plurality of chokepoints creates potential-though far more limited and conditional-vulnerability to pressure or disruption.

Most significantly, both oil and compute are dual-use resources underpinning civilian economies and modern military powers, with disruptions constituting major national security threats. Moreover, countries could use advanced chips allocated for civilian use to jump-start military programs or to strengthen authoritarian governance that exploits AI capabilities.

Just as Middle East oil reserves centered the region in U.S. foreign policy as critical to stable, accessible energy markets, compute could elevate Gulf nations to the forefront of U.S. AI policy. In an age of net U.S. oil exports and a shift from petro to technological and military investments, a new form of strategic coupling is emerging: compute interdependencies between the United States and Gulf partners, notably Saudi Arabia and the UAE.

Should compute become as strategically significant as oil, the U.S.-Gulf relationship stands to evolve into one of asymmetric interdependence, in which Washington relies on the Gulf to enable rapid AI scaling, while in turn providing the high-end technology essential for Gulf states' economic diversification strategies. This shift highlights the national security and geopolitical implications of emerging U.S.-Gulf compute interdependencies and the expanding footprint of U.S.-origin AI technology in the Middle East.

Central to AI scaling is compute capacity, concentrated in large data centers that house sophisticated, energy-intensive chips used for both training new AI models and inference. As AI workloads proliferate, data centers have become a major and rapidly growing source of power demand: In 2023, data centers accounted for 4.4 percent of the United States' annual energy consumption, a figure projected to triple by 2028. This trajectory is colliding with domestic permitting constraints and grid bottlenecks, increasing domestic pressure to locate compute abroad. Gulf states-particularly Saudi Arabia and the UAE-offer abundant, low-cost energy, making them increasingly indispensable to global AI expansion rather than merely attractive hosts.

Regulatory and geographic factors further reinforce their importance. Compared to the United States, Gulf states host a more accommodating regulatory environment. Proposals such as Emirati and Saudi "digital embassies" allow foreign governments and firms to control and protect data, systems, and policies under their own national laws regardless of infrastructure location, enabling secure deployment of AI abroad and bypassing tedious regulatory hurdles.

At the same time, the Gulf's geographic centrality positions it as a hub to distribute AI services and technology across the Middle East and Africa-an AI gateway to the Global South. For example, the UAE recently committed to a $1 billion investment to expand AI infrastructure and services across Africa, alongside a similar deal led by Emirati AI company G42 in Vietnam. In this context, AI investments provide a channel for U.S. soft power amid declining foreign assistance. As U.S. firms expand infrastructure in the Gulf, they are also expanding the reach of U.S.-origin technology into regions with nascent digital environments, projecting U.S. tech influence in regions where China has had the upper hand.

A critical question going forward revolves around which partners the Gulf will rely on to fill this dependency gap. U.S. firms remain in the lead in cutting-edge semiconductors, but this technology is subject to stringent export controls, complicating and protracting potential deals. China, while offering less advanced technology, does not tightly regulate chip exports and leads in open-source AI models, facilitating Gulf states looking to develop local AI infrastructure and platforms at speed. This dynamic highlights another arena for U.S.-China competition in which Gulf states continue to hedge, even amid signals of U.S. alignment, such as the UAE AI company G42's divestment from Chinese technology following U.S. pressure.

For Saudi Arabia and the UAE in particular, China remains a viable alternative to the United States in AI platform and infrastructure development. AI technology offers an opportunity to increase U.S. soft power influence within the Gulf; building and spreading new AI models deepens Middle East partnerships while creating dependence on U.S. technology instead of China's.

Against this backdrop, current U.S. policy seeks to concentrate AI power domestically while also extending the U.S. AI stack globally through trusted partners. Released last July, America's AI Action Plan defines AI power as domestic infrastructure development, scientific innovation, rapid adoption, AI governance leadership, and control over the supply chain. Because projecting AI power requires global compute scale and ecosystem reach, the U.S. AI strategy will increasingly rely on partners such as Gulf states.

The plan also calls for diffusing the full U.S. AI stack to ensure the United States retains the largest AI ecosystem, setting global standards and spreading influence. In line with this, the Trump administration rescinded former President Joe Biden's AI Diffusion Rule last year, facilitating U.S. firms' investment in AI infrastructure abroad. Gulf partners represent deployment nodes in this system.

Given the Gulf Cooperation Council (GCC) members' vast capital reserves and strategic push to diversify their oil-based economies, these conditions create a synergistic environment for U.S. AI investment: U.S. firms rely on Gulf energy, capital, and deployment, while Gulf states depend on technology and platform access to secure their economic transformation.

Interdependence, however, involves a series of risks and tradeoffs. Most consequentially, the United States diffuses AI technology into the Gulf to ensure sectoral dominance and broader influence, but creates dependence risks when infrastructure sits abroad. In previous conflicts, regional adversaries such as Iran and its proxies targeted pipelines, refineries, and oil fields in Gulf partner states. In the compute era, these actors could also target data centers, energy infrastructure supporting compute, and fiber chokepoints such as the Bab al-Mandab Strait off the coast of Yemen, which hosts critical undersea data cables linking internet connectivity in Asia and Europe.

As AI investments and technology proliferate in the Middle East, the high concentration of infrastructure and support systems presents security risks. In 2024, four undersea cables in the Red Sea were cut, disrupting roughly a quarter of internet traffic between Asia, Europe, and the Middle East. Yemen's Houthis were accused of severing the cables, though the group denies these allegations. The high concentration of undersea cables in the strait is subject to accidental damage from anchor dragging and intentional sabotage.

Likewise, the 2019 attack on Saudi Aramco facilities demonstrated that strategically critical infrastructure is vulnerable to regional spillover of conflicts. Drone strikes-claimed by the Houthis-on two oil processing facilities disrupted nearly half of Saudi Arabia's production. As compute becomes a central economic resource, the concentration of data center infrastructure in the Gulf inherits similar exposure. Data centers are susceptible to several additional attack vectors, including network disruption, data theft, application functionality disruption, and physical sabotage of supporting infrastructure such as energy facilities, complicating effective security frameworks.

Apart from kinetic security and concentration risks, U.S. policymakers are right to be concerned about the diversion of advanced semiconductors to China, considering Beijing's technology ties with Abu Dhabi and Riyadh. However, even with the rescission of Biden-era controls, approvals to export advanced Nvidia Blackwell chips to Emirati G42 and Saudi Humain in late 2025 were subject to both companies meeting strict security and reporting requirements.

Offshoring compute capacity may complicate control over U.S. AI governance standards. America's AI Action Plan and the National Security Strategy (NSS) explicitly highlight the importance of leading in global standard-setting for advanced technologies to ensure AI systems reflect U.S. priorities and standards. Yet as the United States exports technology abroad, Washington necessarily cedes some leverage over governance and standard setting. For example, the UAE already represents a pioneer in smart city AI adoption. In late 2024, Abu Dhabi launched the AI platform TAMM-a collaboration between G42, Microsoft Azure, and OpenAI-designed to digitize access to key government services such as healthcare, license renewals, car registrations, and transactions. The presence of two U.S. technology firms highlights a key aspect of this risk: Even if the United States controls the technology and platforms enabling AI deployment, other nations control governance and standard setting.

AI systems also have the potential to enhance authoritarian governance. The UAE's smart city initiatives strive to adopt AI across all aspects of life, including security and law enforcement. "Falcon Eye" and "Oyoon"-eyes in Arabic-are AI surveillance systems used respectively in Abu Dhabi and Dubai that access thousands of cameras to predict and analyze criminal activity through facial recognition and behavior interpretation software. These systems assist police in addressing everything from traffic violations to national security concerns. Yet in authoritarian Gulf states with severe punishment for political dissent, AI-enhanced surveillance has the potential to further restrict political discourse, civil society, and freedom of expression.

Because compute infrastructure creates new vulnerabilities and, therefore, strategic dependencies, current U.S. policy risks underestimating the Middle East's persistent importance. The 2025 NSS emphasizes the region's declining significance relative to other theaters, citing net oil exports and the shift from superpower competition to "great-power jockeying." While the NSS correctly shifts focus to economic cooperation and conflict management, emerging compute interdependence and subsequent infrastructure vulnerability alter the strategic calculus, potentially keeping the Middle East central to U.S. interests.

Increased economic engagement-as emphasized in the NSS-requires sustained conflict management to protect critical infrastructure and assets that were at risk even before data centers. Growing AI interdependence is part of wider economic investments reaffirming the Middle East's position as a key area of focus for Washington. Moreover, with conflict once again looming in the Middle East, the Trump administration should also consider the national security implications of attacks on the Gulf's nascent AI infrastructure among potential spillover impacts of war with Iran.

The United States will struggle to win the global AI race without deepening exposure in a volatile region it officially seeks to deprioritize. Moreover, Gulf states will continue to be central to the landscape of global computing power with or without the United States, creating incentives for Washington to deepen Middle East partnerships. U.S. policymakers should therefore prepare for a new Middle East-focused national security priority centered on compute, data centers, and AI interdependencies. If the United States truly believes compute is the new oil, it should recognize that, like oil, this strategic resource will bind Washington to the Gulf much more tightly than current doctrine acknowledges.

Mona Yacoubian is director and senior adviser of the Middle East Program at the Center for Strategic and International Studies (CSIS). Sam Zabin is an intern with the Middle East Program at CSIS.

Programs & Projects

Commentary by Michael Ratney - January 5, 2026