Russian Servers Went Dark. Did the Right-Wing Internet Go Quiet?

The idea that Ukrainian drones have pulled the plug on Western political bot farms is satisfying. Proving it takes more than a traffic chart.

I would love to believe this story.

There is something enormously satisfying about the suggestion that Ukraine hits a few Russian data centres and, suddenly, the Western far right loses its voice. All that supposedly spontaneous outrage, all those accounts declaring themselves patriots, silenced by a power failure somewhere outside Moscow.

It offers a wonderfully simple explanation for an exhausting political landscape: Russian bots, Cambridge Analytica-style manipulation, and manufactured anger masquerading as public opinion.

But that is precisely why the story deserves scrutiny. Wanting something to be true is a good reason to check it carefully.

The attacks themselves are documented. Ukrainian drones struck Yandex data centres in Sasovo and Kaluga on 8 and 9 October 2026, according to Reuters reporting summarised by Ars Technica. Yandex acknowledged disruption to its services. On 11 October, AP reported another attack shutting down a Yandex facility in Vladimir. There is a substantial infrastructure story here before we get anywhere near Western politics. Ars Technica; Associated Press

Then came the political interpretation.

One LinkedIn post asserted that activity around Germany’s AfD had plummeted after the attacks, pointing to Cloudflare charts showing a fall in automated traffic from Yandex’s network. Yet its suggestion that Kremlin troll farms had been affected was conditional. No accompanying dataset measured AfD accounts, posts or engagement before and after the strikes. Original post

This is where the argument gets ahead of the evidence.

To most people reading a political story, a “bot” means a fake account pushing propaganda. To Cloudflare, it means traffic classified as automated—including the robots Yandex uses to collect pages for its search engine. A fall in that activity could produce an impressive graph without a single political account disappearing. The chart in question covers a Yandex network, AS13238; it does not measure right-wing opinion on social media. Cloudflare documentation; Yandex documentation; Cloudflare Radar

Tom’s Hardware reported that bot traffic from this network fell to 0.4% of its maximum over the preceding two weeks. That is a striking reported decline, although the underlying historical figures have not been independently reproduced for this article. It cannot be translated into a percentage of Russian propaganda accounts taken offline. Tom’s Hardware

None of this makes Russian influence operations imaginary. In July 2024, the US Justice Department announced action involving 968 social media accounts associated with an alleged Russian government-operated bot farm. Investigators described fictitious profiles using AI-assisted content to advance Russian government objectives. X suspended identified accounts. That case involved accounts, infrastructure and investigative evidence—the sort of material needed to move beyond suspicion. US Justice Department

So it is reasonable to ask whether the Yandex strikes disrupted similar operations. It is premature to say the traffic charts have answered that question.

There is also a larger analytical trap here. It is worth asking how much of the online right’s apparent support is manufactured. But assuming Moscow explains it all would be politically convenient. It would allow us to treat domestic movements as an imported infection and avoid examining the institutions, interests and choices that sustain them at home.

We should be just as willing to ask whether a state is manipulating its own population as whether a foreign state is interfering. What roles do domestic political parties, campaign consultants, donors, media owners and platform operators play? Those questions require their own evidence. Ordinary persuasion, selective presentation, covert influence and outright deception are different things.

“Cambridge Analytica-style manipulation” captures a suspicion about how public opinion can be engineered. It is not, by itself, an explanation for this episode. We still have to establish who did what, through which mechanism, and with what measurable effect.

Foreign interference and domestic political agency can coexist. Russian operators might amplify a grievance they did not create. Domestic actors might benefit from that amplification without being directed by Russia. And people can sincerely hold views that others find objectionable. An analysis that cannot accommodate those possibilities will miss much of what it is trying to explain.

What would make the current claim convincing? Start with a defined group of accounts and their posting history. Show that their activity fell unusually after the outages, compared with their normal patterns and other political communities. Establish whether they were suspended, whether the data collection failed, or whether the news cycle simply moved on. Then look for evidence connecting those accounts to the damaged infrastructure.

If the same identifiable network repeatedly stopped when facilities went down and resumed when service returned, that would deserve serious attention. Verified operational links would make the case stronger still.

As of 12 October, the public sources reviewed for this article do not provide that chain of evidence. They establish attacks and report a sharp decline in automated Yandex traffic. They do not establish that those attacks caused a collapse in pro-MAGA or pro-AfD activity.

I would welcome credible evidence that a machinery of political deception had suffered a serious setback. But until the figures are in, the honest position is that we do not know. Russia can amplify a political movement without having invented it. And exposing foreign interference does not absolve domestic actors of responsibility for what they choose to promote.