By Tom Sharpe*
This month (September 2026), Iran’s Revolutionary Guard announced it had recovered an American uncrewed submarine at the entrance to the Strait of Hormuz and released video of it being hoisted out of the water. They called it “one of the most modern, intelligent and unmanned submarines of the terrorist American army”. The Pentagon said it malfunctioned more than a day earlier while “surveying regional waters”, was an older model and carried nothing classified. Both can be true.
The vehicle has been identified from the pictures as an Anduril Dive-LD: 19ft long, three tonnes, roughly 10 days’ endurance and about $2.5m each. The US navy took delivery of its first in April 2025 and has said all along they are “attritable” – a military word meaning “we don’t mind too much if we lose it” – lying on the spectrum somewhere between “disposable” and “exquisite”.
What was it doing? This was most likely mine-detection or more general acoustic surveillance operations. Right now, there is overlap. Mines keep being mentioned and whilst I still maintain they have not been used yet (because Iran will keep them back for a rainy day) the risk remains and therefore having something in the water actively looking to reassure shipping is useful. And besides, much of mine warfare is about eliminating mine-like objects in advance, along pre-surveyed routes, so that if a real mine turns up, it stands out. I spent many a long month doing just this in the crewed minehunting ships that we are now getting rid of.
In years to come, Dive-LDs or similar unmanned subs could carry their own mine-disposal capability, but for now, they would be more likely to be used to cue in an uncrewed surface vessel that could acoustically or magnetically activate the mine using towed “influence sweep” equipment. Alternatively it might cue in a different type of unmanned sub, likely controlled from the surface by wire, able to destroy the mine with shaped-charge weapons.
If you are really interested in a mine you can send down naval Clearance Divers to render it safe and recover it for further examination. But this is time consuming and very high risk, especially when operating within the Iranian weapons envelope. I am minded to suggest that current reporting on how this has been done extensively by US navy SEALs in recent months is incorrect, the more so as such work would be far more likely to be done by US navy Explosive Ordnance Disposal (EOD) divers – the American equivalent of our Clearance Divers.
Likewise, using the Dive-LD to chart the seabed for navigational safety is unlikely. There are few parts of the world better charted than the Strait of Hormuz.
Regardless of what it was doing, a relatively new bit of kit has now been captured by the enemy. Does this matter? Technologically, yes. Iran will take it apart and will pass what they find to Russia or China. They may even hand it over for detailed analysis.
Analysing its acoustics will be useful too. These things are whisper quiet but they will emit noise on certain narrowband frequencies – anything with moving parts does. A fundamental principle of anti-submarine warfare is that if you know what those frequencies are, you can find them more easily. Clearly this situation is nothing like as useful as detecting and exploiting the signature from a nuclear-powered attack or deterrent submarine, but it all helps.
Then there is the data held within the vessel itself that it has obtained from its surroundings. Plus there’s the programming that allowed it to operate autonomously in the first place. Here the Pentagon has been clear – “nothing of use” has been lost. Of course, they would say that, but either way, this is where things in the world of underwater unmanned technology, and the risk of capture, get really interesting.
Communicating through water over any but the shortest distances is basically impossible. Laser technology may overcome this one day, but not yet. So, by the time you have come to a depth where your data can be transmitted – either surfaced or pretty close to it – then returned to depth to do something about it, that information is time expired. In other words, uncrewed underwater systems are very useful as sensors, but much less so as a weapons platform. The way around this would be to make the system genuinely autonomous, but to do this it needs a brain that is able to cut through the nearly unlimited ambiguity of underwater warfare and make the sort of decisions that would normally be taken by a highly trained crew. If the decision is whether or not to use a weapon against a target then a) your system has now removed the human from the kill-chain – something that we are not ready to do just yet – and b) you need to load its brain with the sort of classified mission library that would genuinely be a disaster if captured.
That is jumping ahead, but using vessels like the Dive-LD in an active combat zone will be providing invaluable data to those developing this way of conducting operations. The same will apply to us Brits if the threat ever allows our Royal Fleet Auxiliary vessel Lyme Bay to get close enough to the Strait to deploy her various uncrewed systems. Our versions of all this sit all along the size scale: from hand-launched mine-hunting vehicles a couple of metres long, up to the 12m XV Excalibur, handed over in January and now on trials.
Uncrewed underwater vessels, both large and small, also form a key part of the Atlantic Bastion concept and the Hybrid Navy concept sees one of its four types of unmanned “sloop” – the Type 93 – being an underwater vehicle. The others are a missile barge (T91), an underwater sensing platform (T92) and the air-threat sensor vessel (T94). The vision sees all of these eventually being controlled by the Common Combat Vessel (CCV). It’s an excellent vision, but all the technical (and ethical) challenges highlighted here, and many more, will need to be overcome along the way. I hope we collaborate very closely with the US on all of this going forward.
For now, the Americans have been the first to stumble into one of the hazards. This is not surprising when you consider how far ahead they are and how keen they are to use these new systems, for real, “up-threat”. This is a good example of the “fail fast” culture that moves things forward and is something we need to embrace within our risk-averse defence procurement world. To the Americans this capture will be annoying and a little embarrassing, but not much more. If we compare this to the embarrassment of the Iranians capturing sailors and marines – as we in Britain know only too well – the advantage of uncrewed becomes clear.
This article was first published in The Telegraph and is republished with permission of the author.
*Tom Sharpe OBE served for 27 years as a Royal Navy officer, commanding four different warships
