HYBRID ATTACKS / EUROPE

SABOTAGE IN EUROPE.
CABLES, RAIL, ARSON AND PLOTS.

Reported sabotage and hybrid attacks in Europe and the UK from the last 90 days: undersea cables and pipelines, railways, arson, parcel bombs and plots, and arrests for them. Each incident links to its news reports. For attacks on computer systems, see Cyber attacks; for Russian ships and jets near the UK, see Russian military near the UK.

01 / IN NUMBERS

The last 90 days

LAST 90 DAYS

11

Incidents reported, one card each

LAST 7 DAYS

11

Incidents with a new report this week

COUNTRIES

5

Romania, Sweden, Germany, Lithuania

NEWS REPORTS

12

Checked by an AI model, each linked to its publisher

02 / NEWEST FIRST

Every incident, newest first

03 / READ THIS FIRST

What this page is and isn’t

Sabotage in Europe: common questions

What counts as sabotage on this page?

A reported act or attempt of sabotage, arson, an explosive device or damage to infrastructure in Europe or the UK that the authorities or the report treat as possible sabotage or a hybrid attack, and arrests or charges for one. Cyber attacks, drone sightings and war damage inside Ukraine or Russia are left out.

Is Russia behind these attacks?

Where a report says who is suspected, the card shows that claim in the report’s own words, such as “Russia, suspected by police”. Many cases are never attributed, and a suspicion is not proof.

What is a hybrid attack?

A hostile act short of open war, such as sabotage, arson, cutting cables or paying criminals to carry out attacks, that a state can deny. European governments have warned of a rise in such attacks since 2022.

Have undersea cables been cut in the Baltic Sea?

Several cables and pipelines in the Baltic Sea have been damaged since 2023, often by ships dragging anchors. Investigations into some are continuing; any reported in the last 90 days are listed above.

How is this tracker made?

Every five minutes the server reads news searches in five languages. An AI model (Claude Haiku) keeps only reports of real incidents, translates the headline and names the target and place. Reports on the same target are grouped into one incident. The AI can be wrong, so every report links to its publisher.