Iran's Undersea Cable Threat: Slowdown, Not Blackout

Iran just told the world where to aim if it wants to hurt the Gulf's digital economy: the seabed. Two senior Iranian security officials told Reuters that Tehran has drawn up detailed plans to hit undersea communications cables in the Gulf if the US and Israel strike again, and, more worryingly, that it would try to stop repairs. That second part is the real threat. Cables break all the time. Cables that break and cannot be fixed are a different problem.

Cue the headlines about the internet going dark. It won't. But "no global blackout" is not the same as "nothing to worry about," especially if you run a business in the Gulf or in India. Here is what the threat actually is, what a cable war would look like, and who should be nervous.

What Iran actually threatened

According to Reuters, as published by CNA, the two officials gave few details on how or where an attack would happen. One said Iran would "aim to prevent repairs after any attacks to prolong disruptions to communications, including internet connections." The same official put it bluntly: "If Iran is attacked, there will be no more red lines... from undersea cables to military bases."

That followed a public warning on October 21 from Yadollah Javani, deputy head of political affairs for the Revolutionary Guards (IRGC), that Iran could target seven major internet cables beneath the Strait of Hormuz if Iranian infrastructure came under attack. Another senior figure, Mohsen Rezaei, has said US-linked digital infrastructure across the Gulf, including data centres and AI facilities, could be fair game in a renewed conflict.

So this is not a vague "we could do bad things" line. It is a named target set, a stated repair-denial strategy, and an explicit link to the Gulf's data-centre boom.

The cables under the Strait of Hormuz

As India Today lays out, citing TeleGeography, six active submarine cable systems run beneath the Strait of Hormuz:

• 2Africa
• AAE-1 (Asia-Africa-Europe 1)
• FALCON
• Gulf Bridge International (GBI)
• Tata TGN-Gulf
• OMRAN/EPEG

Two more, FIG and SEA-ME-WE 6, are under construction and expected around 2027. These systems link Gulf states to Asia, Africa and Europe.

Around 99% of international data travels over submarine cables. Yet less than 1% of global international bandwidth passes through the Strait of Hormuz, according to Alan Mauldin, research director at TeleGeography.

Read that twice. Hormuz is a choke point for oil, not for the world's internet. The global web will barely notice a cut there. The countries plugged directly into those cables will notice a great deal.

Blackout scenarios: what a cable war really looks like

Forget the "internet dies" fantasy. Here is how it would most likely play out, from mild to ugly.

Scenario 1: one cable cut, repairs allowed

This happens somewhere in the world every week. TeleGeography notes that on average two cables suffer faults globally each week, mostly from anchors and fishing. Operators shift traffic to other systems, a few routes get slower for a while, and most users never find out.

Scenario 2: several cables cut, repairs delayed

Now the spare capacity starts running out. Traffic has to be rerouted over longer paths, via the Red Sea or east through Singapore. Latency goes up, video calls stutter, overseas sites load slower and telecom operators pay more for transit. Reuters reports analysts warning that multiple cuts could slow internet services and disrupt financial transactions, e-commerce and cloud computing, with the damage depending on how much spare capacity exists on alternative routes.

Scenario 3: multi-cable cuts and repairs blocked

This is the scenario Iran is actually describing. Repairing a subsea cable needs a specialist ship, permits to work in territorial waters, and insurers willing to cover it. In a war zone, all three get harder. Reuters notes that higher insurance costs and permit delays could further complicate repairs. The result is not a dead internet. It is prolonged congestion: weeks or longer of degraded, expensive, laggy international connectivity for the countries that depend on these routes.

We have seen this movie before: the Red Sea

The best guide to a Hormuz cable crisis is what already happened one choke point over.

February 2024. Three Red Sea cables, AAE-1, EIG and Seacom/TGN-Eurasia, were damaged near Bab el-Mandeb. The likely cause, per the International Cable Protection Committee as reported by gCaptain, was the dragged anchor of the Rubymar, a cargo ship abandoned after a Houthi missile strike. Repairs depended on getting permits to work in Yemeni waters, and the fixes were not completed until mid-2024. Users across East Africa and Asia lost capacity. Nobody lost the internet.

September 2025. Failures on the SMW4 and IMEWE systems near Jeddah degraded connectivity in India, Pakistan, the UAE and Saudi Arabia, according to NetBlocks, with slow speeds and intermittent access. Microsoft warned Azure traffic through the Middle East could see higher latency.

Notice the pattern. Both incidents were accidents or side effects of war, not a deliberate campaign, and both still caused months of headaches. A planned strike on several Hormuz cables, with repair crews kept out, would be that pattern on purpose. Reuters also notes the Red Sea could be a more significant target than Hormuz if Iran's Houthi allies joined in, although Tehran has said nothing publicly about that.

What it means for India

India is not in the Gulf, but its internet runs through it. Cables come ashore at landing stations in Mumbai and Chennai, and India Today's OSINT work found that at least four of the seven Hormuz-linked systems have Indian links. Three are live: FALCON, GBI and AAE-1. The fourth, SeaMeWe-6, has been landed in Mumbai and Chennai by Bharti Airtel but is not due in service until 2027. Airtel is in the SeaMeWe-6 consortium; Reliance Jio is a member of the AAE-1 consortium.

The exposure numbers vary by how you count. An Economic Times estimate cited by India Today puts roughly a third of India's westbound traffic through the Hormuz corridor. Mint has estimated that about 60% of India's international traffic runs via Mumbai and West Asian routes, which bundles the Red Sea and Hormuz together. Either way, a simultaneous hit on both choke points is India's worst case.

Who feels it?

• Everyday users: slower overseas websites, laggy video calls, buffering on services hosted abroad. Not a dead phone.
• UPI: largely unaffected. Domestic payments run on domestic rails.
• IT services and BPO: the real exposure. Client calls, data transfers and remote work with the US, Europe and the Gulf all depend on stable international links.
• Banks: cross-border transactions and links to overseas systems could see delays.
• Telcos: rerouting via Singapore works, but it is longer, slower and more expensive.

The Gulf has more to lose

For the Gulf states, this is not a latency problem. It is a strategy problem. They have spent heavily to become data-centre and AI hubs, and Iran has now openly named that infrastructure as a target. States with fewer route options are the most exposed. Some have terrestrial alternatives over land, and Saudi Arabia is less dependent on Persian Gulf routes because it also has landings on the Red Sea. That spreads the risk; it does not remove it.

The Eptune take: plan for choke points, not apocalypse

The lesson here is not "panic." It is that geography is a single point of failure, and most companies have never mapped theirs. If your product, support team or cloud region quietly depends on one corridor, you have a resilience problem whether or not Iran follows through.

What smart teams should do now:

• Know which cloud regions your critical workloads and customers use, and what happens if Middle East routes slow down.
• Ask your telcos and cloud providers about route diversity. "Multiple cables" means little if they all pass through the same strait.
• Design for degraded mode: async workflows, cached content, and fallback regions, not just for outages.
• Treat infrastructure risk like security risk. We made the same argument in why security vulnerabilities are a growth problem, not an IT problem.
• If your stack leans on cloud-hosted AI models, latency spikes hit you first. Our November 2026 AI model rankings are worth reading with that lens: where a model runs matters as much as how it scores.

Iran's threat is a reminder that the cloud is not in the sky. It sits on the seabed, in a handful of narrow straits, and somebody just said out loud that they know exactly where. For more on how tech and strategy collide, browse the Eptune blog.