The Iran Ceasefire Might Stabilise the Middle East — But Can the AI Boom Recover?

April 2026

IoT & Emerging Technology

Lorien Carter

Senior Research Analyst

Tanker traffic through the Strait of Hormuz has effectively ground to a halt, down from an average of 138 vessels per day in peacetime. The impact on oil markets has been immediate and severe; marking the most significant price shock since the 1970s, with further volatility likely as the conflict evolves. Prior to US strikes on Iran on 26 February, oil was trading at $74 per barrel; it has since surged, hitting peaks of $113.

Crude Oil, $ per Barrel (as of 9th April)

This week’s ceasefire has provided only limited relief. Although prices briefly dipped following the announcement, the truce remains fragile, and shipping through the Strait is still heavily restricted; meaning supply constraints, and elevated prices, are likely to persist in the near term.

This disruption is not confined to oil markets. It feeds directly into the cost and availability of energy inputs that underpin AI infrastructure; particularly natural gas, which remains the primary power source for data centres.

Around 75% of AI data centres run on natural gas as their primary fuel source, as outlined last month by Cleanview. This is because of unavoidable baseline energy demands: local grids cannot provide AI data centres with enough power, and even if they could, connecting to them can take up to five years.

Additionally, solar power and other renewable sources experience fluctuating energy production, while small nuclear reactors are up to a decade away from being realised in a commercial space. The only method deployable today that is capable of generating enough power for AI is natural gas; making it the foundation of the entire AI sector. As a result, any increase in oil prices or limiting of supply represents a material threat to the AI industry — a risk already highlighted by Iran’s attack on Qatar’s Ras Laffan Industrial City, the world’s largest liquefied natural gas facility.

Electricity accounts for around half the expenses of an AI data centre. With the sharp surges in energy prices affecting bottom-line expenses, it will become even harder for AI startups to become profitable. Downstream effects such as increased subscription prices or usage caps might reduce demand for AI subscriptions, as well as causing investors to think twice before supporting the AI industry.

Helium Shortages

At the start of the year, we wrote about the deleterious effects of AI on consumer chip prices. In short, AI hyperscalers were outbidding consumer manufacturers for access to chips; causing prices to soar and smaller AI startups to be locked out of RAM access - now, AI and computer chips are in the headlines again. 

Helium is critical to the process of building advanced microchips for AI. A third of all commercial helium is produced in just three helium plants, located in Qatar. The immediate effect of the closing of the Strait of Hormuz is that helium tanks are stuck in Qatar, unable to transit. At the moment, the impact of this on supply is mitigated by the recent 15% surplus of helium compared to global demand. However, if the war continues then these supplies will deplete. Furthermore, as helium is produced from waste products of liquified natural gas (LNG) plants, when these plants have to cease production - when their LNG storage tanks are filled - it will have a knock-on effect on helium production. At worst, this could close a third of helium production; devastating semiconductor manufacturers and, once again, increasing prices.

Threats on AI Data Centres

For the first time ever, commercial data centres have become targets for warfare. Several data centres in the Middle East region have been struck by missiles: Amazon Web Services has had data centres struck in Bahrain and Oracle has had one hit in Dubai. These strikes had the effect of wiping out access to mobile banking services and fintech-based payments for millions in the region.

What’s more, Iran has stated that if the US hits civilian infrastructure such as power plants and water desalination plants, they will retaliate with strikes against the energy and technology infrastructure of US companies - specifically, the Stargate data centre in the UAE, which is jointly owned by OpenAI, SoftBank, and Oracle. This unprecedented move recognises AI infrastructure as a sovereign asset.

Regardless of how quickly this current war ends, the positioning of data centres as valuable targets will not be forgotten. It costs billions to build one single data centre capable of meeting the requirements needed to power AI, and only seconds to completely wipe it out. This reliance on exposed core infrastructure may cause investors to pull back.

What of the Future?

No matter how quickly the war ends, ongoing shocks to supply will continue. Existing attacks on oil production plants have eliminated production capacity for years to come. For example, attacks on the LNG plant in Ras Laffan has been estimated to wipe out 3.5% of global LNG production for the next three to five years. This will increase energy costs in the long term; impacting AI’s supply chain and operating costs.

One potential long-term effect stems from the geographical divergence in AI growth, which could result in regions such as North America and South Asia racing forward with AI while Europe and the Middle East are held back. Europe is more exposed to oil price shocks than the US - seeing prices rise 40% compared to only 5% in the US - so oil price rises will hit Europe-based AI companies harder. Similarly, the conflict risk to infrastructure may cause Chinese companies such as Alibaba and Huawei to redirect data centre investment away from the Middle East; interrupting the Gulf’s ambition to become an AI frontrunner.

While the US is initially more insulated from oil price rises, the long-term economic impacts of the war may yet put its entire AI industry at risk.