The Unexpected Effect on AI from the War on Iran

On February 28, 2026, US and Israeli forces struck Iran. Most stories initially focused on the geopolitics and geoeconomics implications and the fallout from higher gasoline prices. However, within 72 hours, Iranian drones hit three AWS data centers in the UAE and Bahrain — the first confirmed military attacks on commercial cloud infrastructure in history. Within days, the Gulf states which are funding a large part of the global AI buildout quietly started reviewing contracts and whether they could walk away, reduce, or delay their commitments (Reuters). The implications of this will have a ripple effect on AI companies and the U.S economy as a whole.  Over 90% of GDP growth in the U.S (H1 ‘25) was from AI investments (Fortune). A third of the S&P market cap is from the Magnificent Seven which are riding the AI boom! (Yahoo News)

How Gulf Money Became Critical

Gulf states have been investing heavily in US AI through sovereign wealth funds. Roughly $90B of AI investments are committed with up to $300B promised over multi-years. While many focus on the size of the investment, what’s less understood is the role this capital has been playing inside the AI Ecosystem. The AI buildout has been constrained along three dimensions:

• Power. U.S. data center expansion is increasingly limited by grid capacity, transformer shortages, and permitting timelines. Some utilities have already slowed or halted new connections. Gulf markets offer abundant energy and faster deployment cycles. Morgan Stanley projects a 49GW power shortfall by 2028 — roughly Canada’s entire grid demand.
• Capital. Hyperscalers are funding infrastructure at unprecedented levels and approaching practical limits of balance sheet expansion. OpenAI alone is projected to require over $200B in compute by 2030. Gulf funds bring long-duration capital without near-term return pressure
• Market Substitution. Export controls have reduced access to China. Gulf partnerships provide an alternative source of demand and deployment, partially offsetting that gap. As the Stimson Center noted, these deals “generate revenue that offsets losses from economic decoupling with China.”

What Actually Got Funded

Before the war, capital was already being deployed at scale (Mubadala Annual Review 2024; QIA press release; Arab News):

• HUMAIN (Saudi Arabia’s state AI company): $23B in signed deals with Nvidia, AMD, AWS, and Qualcomm
• MGX (Abu Dhabi): Co-led a $30B AI infrastructure fund with Microsoft and BlackRock; invested in OpenAI and Anthropic; contributed ~$7B to the Stargate project
• QIA (Qatar): Anchored Anthropic’s $13B Series F funding round; co-anchored xAI’s $20B round with MGX
• Stargate UAE: Planned 5GW AI campus in Abu Dhabi — the largest outside the US
• Mubadala: Owns 89% of GlobalFoundries, the only major chip fab on US soil that isn’t TSMC

The Assumption That no Longer Hold

The security framework built around all of this is called Pax Silica. 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.”  As one analyst put it, these frameworks were designed for geopolitical competition, not for disruption at the infrastructure layer. Data centers were treated as neutral assets. That distinction is less so now given the potential dual-use of these infrastructure (civilian and military) (Rest of World)
The economic fallout is uneven across the Gulf. Saudi Arabia is relatively insulated — it has large reserves and can export oil via a Red Sea pipeline that bypasses the closed Strait of Hormuz. The UAE took direct infrastructure hits. Qatar is most exposed: its main export revenue dropped to near-zero when Hormuz closed. Goldman Sachs estimates Qatar and Kuwait could see GDP fall 14% if the conflict runs to the end-April.

If Gulf Funding Slows, What Might Give

As Gulf countries face fiscal challenges due to revenue loss and need to divert funds to rebuild damaged infra and increase defense spending; pressure on AI investments will hit. There is also no equivalent substitute at this scale. Other capital pools are either fully deployed or operate under tighter constraints.  Scenarios that could play out:

• Short conflict:  Temporary delays; projects resume with higher cost structures and embedded risk premiums
• Medium duration: Select cancellations; divergence across Gulf states; meaningful delays in committed capital
• Prolonged conflict: Capital reallocation; smaller funding rounds; multi-year impact on infrastructure pace

The impact would not be immediate or uniform necessarily, but it would surface in stages:

1. Capital markets first: Gulf funds reduce liquid positions in stock market; large-cap tech sees pressure before infrastructure signals appear
2. Funding rounds adjust: Mega-rounds shrink or take longer to close; valuations compress
3. Infrastructure timelines slip: Projects like Stargate UAE delay; regional hubs undergo security and cost reassessment
4. Capacity planning shifts: Hyperscalers re-sequence buildouts; expected compute supply moves further out

What Changes Regardless 

Several shifts are already in motion regardless of how long the conflict last:

• Data centers enter the risk model: Site selection, insurance, and physical security become first-order considerations
• The Gulf diverges internally: Different exposure levels lead to different investment trajectories
• Geographic redistribution: Regions like India, Northern Europe, and Southeast Asia become more competitive for new capacity
• Relative timelines matter more: Any delay in the U.S.-aligned infrastructure expansion extends the window for other competitors to build