They Droned Back - Students investigate ships linked to drone swarms

[jarlrmai]

This article details an investigation by a group of students into drone swarms and links to ships "Russian spy ships"

https://www.digitaldigging.org/p/they-droned-back

Seven German journalism students tracked Russian-crewed freighters lurking off the Dutch and German coast—and connected them to drone swarms over military bases.

It details a lot of OSINT techniques that were apparently used to track apparent Russian spy ships and cross references them to drone sightings.

However they seemed to have skipped a vital step in OSINT, verifying there were actually any rogue operated drones at all..

 

[Mendel]

Why do they call freighters "spy ships"? The Yantar does reconnaissance, and it's equipped for it. Freighters are not.

Russia's only Mediterranean naval base. Where Russian submarines dock.
Coincidence?

Not a coincidence. It's the only Russian port in the Mediterranean, so we expect Russian ships of all types to dock there.
It's not a coincidence, but it's meaningless. Same with the Russian ship that was overhauled in Russia.

According to classified BKA reports obtained by the team: 1,072 incidents involving 1,955 drones in 2025 alone (as of November 19). Forty-five percent occurred in evening hours. Drone swarms flew "almost exclusively over or near military installations."

In only 29 of 498 investigated cases could drone pilots be identified. In none of those cases were they state actors. In 88 percent of cases, authorities couldn't even identify the drone type.

3 "drone incidents" per day, on average. Many sightings in the evenings, when people are off work (and other people fly their hobby drones). And military installations have bored guards filing reports. Missing information: are those military reports also predominantly in the evenings?

Then we have the UFO thing going: all resolved incidents are not malicious, most incidents are so deep in the LIZ that not even a type can be determined, so the logical conclusion is that these are hobbyist drones as well. Instead, those are the cracks where the real scary drones are hiding, that nobody has ever seen close enough.

The team's final tally: they could draw 19 temporal and geographic correlations between drone sightings and the positions of the three ships.

The article mentions these "correlations":
"Distance from the Lauga incident: 115 kilometers. Within drone range."
"Then it started circling—25 kilometers offshore, directly in front of a Belgian military base."
"Seventy kilometers away, the HAV Dolphin had been anchored in the Ems estuary for three days."
So basically, these ships don't pull into port between jobs (because these are Russian shadow fleet, they can't just take any job), and instead circle at sea, presumably to save harbor fees. Missing information: how many other ships do this that are not spy ships?
And a "correlation" is whenever during this time there's a "sinister" drone report in a 150 km range during that time (remember, 3 reports per day). Coincidence? Probably!

A drone with 230+ km range should be big enough to be identifiable and trackable.

Belgium's coastline is 65 km long, Ostende is right in the middle, and probably has a naval base, so the "25 km offshore, directly in front" is meaningless.The ship may have been killing time waiting for a job in Ostende.

And these were their best examples, picked for the presentation!

For context, "Axel Springer Academy for Journalism and Technology" reads the same as "Rupert Murdoch School for Journalism". I don't know anything about them, but the mental association is "boulevard" and "yellow press".

 

[jarlrmai]

Why do they call freighters "spy ships"? The Yantar does reconnaissance, and it's equipped for it. Freighters are not.

Covert military ships operating as commercial freighters/trawlers are a thing and were big in the cold war.

AGI ships

https://en.wikipedia.org/wiki/Spy_ship

Lines get blurred.

 

[Mendel]

Covert military ships operating as commercial freighters/trawlers are a thing and were big in the cold war.

Right. And now they have satellite surveillance, so having "AGI ... assigned to a single patrol station for as long as six months" is no longer needed. And neither is flying drones for hundreds of kilometers to do recon.

(Apologies, I extended my previous post while you weren't looking.)

 

[jarlrmai]

Right. And now they have satellite surveillance, so having "AGI ... assigned to a single patrol station for as long as six months" is no longer needed. And neither is flying drones for hundreds of kilometers to do recon.

(Apologies, I extended my previous post while you weren't looking.)

Ah yeah your edited post makes it clearer what you were getting at.

 

[stevewhite]

The "We Droned Back" part, the part they chose to highlight as the headline is extremely underwhelming

They flew a drone over one of the freighters and found a regular freighter...

They flew their own drone over the HAV Dolphin.

From the air: 88 meters long, 12 meters wide. Open cargo hold with grid covers, typical for multi-purpose freighters.

The full data is behind a paywall.

 

[Mendel]

But here's what the team learned from security sources: those inspections were "superficial" and "symbolic in nature." Not all containers were opened. You can't properly search a freighter like that without many investigators and more time, they were told.

A container for drone deployment needs to be accessible, it probably needs an electrical hookup, or an exhaust, or the drones are fuel-powered and need a small runway. If I was looking for deployable drones, I wouldn't need to touch most of the containers.

 

[Stijn Oerde]

This is some very poor OSINT research in my opinion.

They were looking for anomalies. Cargo ships move from A to B. Efficiency matters. So when the team saw a ship track that looked like someone had scribbled furiously with a purple marker—loops, circles, ten days of chaos in Kiel Bay instead of a clean transit line—the audience laughed at how absurd it looked.

First of all they don't seem to understand that semi circles or circles like this is what an anchored ship looks like on AIS data. It also isn't necessarily weird for a ship to anchor for a while. They could for example have to wait for a spot in a harbor to open up or wait for a new job. It is probably cheaper to anchor at sea for a bit instead of paying money to a harbor.

At approximately 1 AM, the police vessel reports: "Seven drones detected around the deployment ship." The drones circle both the Russian freighter and the German police ship for hours. Three eventually depart. Four keep circling. Eventually, the police vessel breaks off its escort.

The cases of drones being spotted sound very much like cases where confirmation bias can be involved. Especially with the "drones" around a Russian freighter in the middle of the night. The police were send to check on a Russian ship and because there is so much new about Russian drones they can easily mistake stars or planes as drones.

Then they found a 2024 Rosatom presentation. It showed an orange-and-blue drone on the helipad of a massive red icebreaker in an Arctic landscape.

How they connect this story to a specific Rosatom drone is also completely baffling. They found some vague connections of the Russian ships to Rosatom and then find a drone in a random presentation and then conclude that must be the one. Also, if the Russians were spying with drones it likely wouldn't be Rosatom, but the security services. They could of course use a Rosatom drone, but that is pure speculation.

Also, 19 out of 1000 (or 500 depends on whether they included all sightings or only near important infrastructure) drone sightings being within 200 km of an anchored Russian ship doesn't seem a very impressive correlation to me.

This whole thing is written to make it feel exciting and mysterious with the whole chase thing. But in the end they didn't find anything of note. It seems to be pure confirmation bias fueled speculation in my opinion.

 

[gargamel]

I don't know a lot about maritime trade but I do know that ships often anchor and/or circle about, waiting for calls, as "auctions" or spot trade or whatever is ongoing.

 

[Stijn Oerde]

I now did a small investigation using AIS data myself:

I checked the HAV Snapper on globalfishingwatch.org and found the "incident" when you add a scale it is only moving 200-300 meters back and forward. Clearly a case of a ship being anchored.

The last time it visited Russia was in 2023. It also seems a pretty busy ship. Nothing abnormal. You can also see it anchor at sea on multiple other locations.

On vesselfinder.com it can be seen that the HAV snapper was at anchor at the time of the incident at a designated anchor spot as indicated by the green box. Two ships are there right now also making "suspicious patterns".

Near Kiel there is no such "ship parking lot" though. I don't know enough about maritime law to know if it is illegal or suspicious to park your ship outside designated areas.

 

[JMartJr]

First of all they don't seem to understand that semi circles or circles like this is what an anchored ship looks like on AIS data.

I don't know a lot about maritime trade but I do know that ships often anchor and/or circle about, waiting for calls, as "auctions" or spot trade or whatever is ongoing.

Nice catch. I knew that from "anchoring out" on sailing trips and having to keep an eye out to be sure you don't swing over into a marker, or the shore, or somebody on a shorter anchor line... but never would have thought of it in terms of that image... you win 1 Internet!

 

[Mendel]

First of all they don't seem to understand that semi circles or circles like this is what an anchored ship looks like on AIS data.

Ooooo, that's super obvious once you realize it. IF ONLY THEY HAD INCLUDED A SCALE INDICATOR!

Near Kiel there is no such "ship parking lot" though. I don't know enough about maritime law to know if it is illegal or suspicious to park your ship outside designated areas.

You can anchor anywhere except where it is forbidden (because you'd be an obstruction, there's a cable, etc.).

 

[jdog]

Huh, here's the track of Evergreen Mariner, currently off Long Beach, and the Balboa anchored off Astoria:

 

[John J.]

They could for example have to wait for a spot in a harbor to open up or wait for a new job. It is probably cheaper to anchor at sea for a bit instead of paying money to a harbor.

If it's a sanctioned ship, would it- or its crew- be going into harbour in an EU member state?
Would it be able to make use of services from that nation while moored off its coast?

 

[Mendel]

If it's a sanctioned ship, would it- or its crew- be going into harbour in an EU member state?
Would it be able to make use of services from that nation while moored off its coast?

It's not.

Article:

HAV Shipping CEO Petter Kleppan responded to the team's findings: "HAV has exclusively large, established European companies as customers. We transport dry goods from port to port—steel, curbs, grain, scrap. Typical invoice: about €50,000. We have no Russian customers and generate no revenue from Russia. HAV has 'self-sanctioned'—we don't transport goods to or from Russian customers, and we don't work with Russian brokers."

Source: https://www.digitaldigging.org/p/they-droned-back

 

[Mendel]

Let's talk about this photo:

Then they found a 2024 Rosatom presentation. It showed an orange-and-blue drone on the helipad of a massive red icebreaker in an Arctic landscape.

The ship behind the helipad is the АРКТИКА-1, a small container ship:

Source: https://ecoshp.ru/fleet/sudno-a1/ ( https://ecoshp-ru.translate.goog/fleet/sudno-a1/?_x_tr_sl=ru&_x_tr_tl=en&_x_tr_hl=en&_x_tr_pto=wapp )

Doing a reverse image search, I found a press release from the drone manufacturer at https://ai.mipt.ru/news/tpost/ueim3a4ra1-pervii-rossiiskii-arkticheskii-bpla-dlya :

Article:

MIPT, together with the Northern Sea Route Directorate of the Rosatom State Corporation and Irbis Sky Tech LLC, successfully completed a series of flight tests of an operational ice reconnaissance system based on a vertical takeoff and landing unmanned aerial vehicle. This unique Russian system, developed at MIPT, not only provides operational monitoring data, but also provides a local, three-day forecast of ice conditions in a given water area.

The device took off from the deck of the nuclear-powered icebreaker Yamal and collected operational information on the ice conditions in the Gulf of Ob in the Kara Sea. During the flight, radar images were collected and video recordings were made in the optical and infrared bands. All georeferenced data was transmitted to the operator on board. Specialists also tested the automated landing system on the deck without human intervention.

Source: https://ai-mipt-ru.translate.goog/news/tpost/ueim3a4ra1-pervii-rossiiskii-arkticheskii-bpla-dlya?_x_tr_sl=ru&_x_tr_tl=en&_x_tr_hl=en&_x_tr_pto=wapp

The Yamal is indeed massive and red, and the Kara Sea is in the Arctic, but you can't tell either from that photo.

It's highly unlikely that a specialized drone system like this would be used to buzz a German police escort in the North Sea. It's also unlikely that a drone this big can be hidden away on a ship, or not be identified.

Source: http://ai.mipt.ru/news/tpost/7txc97b3z1-deistviya-sozdannogo-v-mfti-palubnogo-be

 

[gargamel]

Apart from how dumb it seems to say "here is a drone taking off from a ship in the Arctic, therefore these alleged sightings must be due to drones from ships in the North Sea", let's look at specifics:

That particular drone, "Irbis 538" is a massive piece, has five propellers (four vertical, one pusher) and a 5 meter wingspan according to the datasheet:
https://irbisskytech.ru/katalog/irbis-538-vtol/

It is quite conspicuous, and with all these propellers and stuff it is about as unstealthy as it gets. You mean to say that this thing, with completely illogical positioning lights and stuff too, just keeps flying around taunting people all over Europe, people who for some reason keep failing to actually track them and find out where they are coming from, and return to, unscathed, time after time?

In Ukraine things of lesser dimensions, and more "stealthy" (as say Geran/Gerbera) are reliably tracked from the moment they enter UA airspace, and shot down by the hundreds, a week. No lights either, obviously.

Help me make it make any sense.

 

[Mendel]

That particular drone, "Irbis 538" is a massive piece, has five propellers (four vertical, one pusher) and a 5 meter wingspan according tohe budding the datasheet:
https://irbisskytech.ru/katalog/irbis-538-vtol/

Flight time: up to 10 hours
Flight range: up to 1000 km

Security and safety : monitoring large areas, including industrial sites and borders.

Great for the budding drone spotter, except that it has an internal combustion engine, so it's hardly stealthy

 

[Stijn Oerde]

The International Institute for Strategic Studies has now released a report going even further than the German students and saying that it is very likely that Russia conducted hundreds of UAV flights over critical from Russian linked ships.
https://www.iiss.org/research-paper/2026/06/russias-uav-campaign-over-europe/

Thid report is, in my opinion, total BS. It is a confirmation bias fueled mess that treats eyewitness accounts as 100% credible, confuses correlation with causation and does not even proof correlation.
It says it is very likely that there were Russian drones because within a range of 300 km of a drone incident a Russian linked ship can always be found. That is first of all a huge range and as there are hundreds of Russian linked ships sailing through Europe it will become inevitable that one will be within 300 km. They didn't include any statistical analysis or negative control, but just state it is a correlation.
There are so many things more wrong with this report that I can go on and on, but I have better things to do.
Wiebe de Jager from dronewatch made a larger analysis about it:
https://www.dronewatch.eu/new-iiss-...t-drone-theory-but-fails-to-provide-evidence/

 

[FatPhil]

The International Institute for Strategic Studies has now released a report going even further than the German students and saying that it is very likely that Russia conducted hundreds of UAV flights over critical from Russian linked ships.
https://www.iiss.org/research-paper/2026/06/russias-uav-campaign-over-europe/

I don't care about low blows, so I'll just point out that their somewhat unconventional "Uninhabited Aerial Vehicles" would refer to all aerial vehicles, as even the ones packed full of people only contain people who don't *inhabit* that vehicle. If they're using that expansion because they are trying to avoid supposedly gendered terminiology ("unmanned"), then that reveals other holes in their suppositions ("man" as a verb and thus "unmanned" as an adjective are no more specifically gendered than the related "mankind" is - i.e., not at all - so no substitution is desirable or necessary).

 

[jarlrmai]

I don't care about low blows, so I'll just point out that their somewhat unconventional "Uninhabited Aerial Vehicles" would refer to all aerial vehicles, as even the ones packed full of people only contain people who don't *inhabit* that vehicle. If they're using that expansion because they are trying to avoid supposedly gendered terminiology ("unmanned"), then that reveals other holes in their suppositions ("man" as a verb and thus "unmanned" as an adjective are no more specifically gendered than the related "mankind" is - i.e., not at all - so no substitution is desirable or necessary).

Could just be a bad translation?

 

[Mendel]

If they're using that expansion because they are trying to avoid supposedly gendered terminiology ("unmanned"),

"uncrewed" exists

 

[Mendel]

The International Institute for Strategic Studies

Article:

The IISS: The Myth and Ethics of Think-Tank Independence

Last week, Bahrain Watch published an investigation into the secret funding of the International Institute for Strategic Studies (IISS), ahead of its annual Manama Dialogue. The investigation demonstrated that one third of the think-tank's funding was coming from the Bahraini ruling family. In this article, the investigative researchers behind these revelations delve into the activities and role that the IISS has played over the past twenty-five years—critically questioning the IISS claim of "independence," and its relationship with the Bahraini and British governments.

IISS claims one of its charitable objectives is to "promote the adoption of sound policies to further global peace and security and maintain civilised international relations."ًً Yet the nature and terms of security propagated by IISS are determined through back channels and private bilateral meetings between authoritarian and imperialist government officials.

Source: https://www.jadaliyya.com/Details/33838

I would expect them to be heavily biased.

 

[Mendel]

Wiebe de Jager from dronewatch made a larger analysis about it:
https://www.dronewatch.eu/new-iiss-...t-drone-theory-but-fails-to-provide-evidence/

Link policy on Metabunk really wants you to quote your sources.

In several cases, multiple vessels are presented as possible launch platforms, while some were located well over 100 kilometres from the reported sightings. These are hypotheses, not verified facts.
[...]
Like earlier investigations such as They Droned Back, the IISS report does not present direct evidence linking individual drones to specific vessels. Instead, it largely infers such links from the coincidence of reported drone sightings and the presence of Russian-linked ships in the wider area. Correlation can be an important investigative starting point, but it is not, by itself, proof of causation.
[...]
Together with Dutch newspaper Trouw, Dronewatch conducted an extensive investigation into 61 reported drone sightings across Europe. Rather than accepting initial statements at face value, each incident was reconstructed using available witness accounts and official reports.

The outcome was sobering.

Many reported drones turned out to be perfectly ordinary aircraft, helicopters, stars, planets or other explainable phenomena. In numerous cases there was simply no evidence that a drone had ever been present.

By contrast, the few confirmed sightings of Russian military drones in Europe involved aircraft that crossed into neighbouring countries bordering Russia or Ukraine, where they were tracked by air defence systems or later recovered. Those verified incidents are fundamentally different from the largely unsubstantiated reports from Western Europe.
[...]
Despite multiple inspections and boardings of suspected shadow fleet vessels, authorities have not publicly reported finding drone launch systems, recovery equipment or operational UAVs aboard these ships.

The IISS report therefore asks readers to accept a significant conclusion based primarily on circumstantial evidence and temporal correlations.

That is not the same as proof.

 

[John J.]

If they're using that expansion because they are trying to avoid supposedly gendered terminiology ("unmanned"), then that reveals other holes in their suppositions ("man" as a verb and thus "unmanned" as an adjective are no more specifically gendered than the related "mankind" is - i.e., not at all - so no substitution is desirable or necessary).

Agree that "uninhabited" is inelegant, I don't know if it's widely used.

Some other manufacturers/ users use variations on "uncrewed",
DARPA use "uncrewed aerial systems" in a press release quoted by Navy News, 25/05/24, elsewhere they also use "unmanned"

DARPA press release
...The goal of ANCILLARY is to increase small vertical take-off and landing (VTOL) uncrewed aerial system (UAS)...

https://www.navalnews.com/naval-new...s-for-ancillary-vtol-uncrewed-aerial-systems/

Airbus, "Uncrewed Aerial Systems" https://www.airbus.com/en/products-services/defence/uas
BAE Systems, "Uncrewed Air Systems", https://www.baesystems.com/en-uk/product/uncrewed-air-systems, elsewhere sometimes "unmanned",
UK Ministry of Defence "uncrewed air systems" https://www.gov.uk/government/publi...-the-uks-approach-to-defence-uncrewed-systems

 

[Mendel]

The International Institute for Strategic Studies has now released a report going even further than the German students and saying that it is very likely that Russia conducted hundreds of UAV flights over critical from Russian linked ships.
https://www.iiss.org/research-paper/2026/06/russias-uav-campaign-over-europe/

Our argument is not that every reported sighting was Russian-directed, or that every reported sighting involved a UAV, but that the aggregate pattern of UAV sightings cannot be adequately explained by misidentification, hobbyist activity or opportunistic harassment alone.

That's the "there are so many reports, they can't all be wrong" argument that we see UFO believers use when none of their reports hold up individually.

My favorite counterexample is the medieval witch hunts: there were so many witches reported (and sometimes tried and found guilty), surely witchcraft must be real!?

People use this type of argument when they don't have good evidence, and it really looks like motivated reasoning: they're trying to find evidence for a desired conclusion (and failing), instead of following the evidence where it leads.