Russian bot farms are still trolling, even after Ukraine’s drone strikes on Yandex

Ukrainian military drones struck several Russian data centers belonging to Yandex, a Russian technology company. The operation mattered because Yandex infrastructure was linked to online services that some observers associated with Russian information activity. Any disruption raised questions about whether Russia’s influence operations had also been damaged.

President Volodymyr Zelenskyy said the campaign made social platforms feel easier to use. He claimed there was less hate, less division, and fewer provocations. Some social-media users made similar claims, including reports of less traffic around MAGA-affiliated and AfD-related accounts, plus fewer spam calls and messages.

The article presents a different picture. It says the strikes appear to have affected Yandex’s web crawlers, not political disinformation operations. Bot traffic tracked by antibot4navalny remained steady, and Kremlin-aligned accounts continued posting false claims after the strikes. The claimed improvement therefore remained unverified.

A bot farm is a coordinated operation that manages many online accounts to make messages appear more popular, widespread, or authentic than they really are. Some accounts may be automated, while others can be controlled by people using shared instructions. The goal can include spreading propaganda, attacking opponents, creating confusion, or pushing selected topics into wider discussion.

The article gives one concrete example. On Monday, Kremlin-aligned accounts posted videos falsely claiming that experts wanted the European Union to lift its ongoing sanctions on Russia. Coordinated accounts can repeat the same idea, share one another’s posts, and target particular communities, making an organized campaign look like spontaneous public opinion.

The article does not provide a precise definition or technical breakdown of the farms. It does report that antibot4navalny, which tracks state-sponsored troll farms, found bot traffic steady after the strikes. That suggests coordinated activity continued even when some users believed it had sharply declined.

The article does not state how many accounts the Russian operations control or how many posts they produce. That missing figure matters because a campaign’s apparent reach can be confused with its real size. A small network can gain attention through repetition, while a larger network can target several communities at once.

In general, bot-farm operations may combine automated accounts, human operators, and accounts that are only partly automated. They can schedule posts, repeat phrases, share selected content, and react quickly to trending topics. Researchers usually measure activity through account behavior and coordinated patterns rather than assuming every visible interaction represents a unique person.

For this article, the safest conclusion is limited. Antibot4navalny reported that bot traffic remained steady after the Yandex strikes, but it did not provide a total number of accounts or posts in the supplied text. Claims about scale therefore require separate evidence, not estimates based on social-media impressions.

The distinction matters because disrupting a company’s technical systems is not the same as disabling a political influence campaign. Web crawlers collect or index online information, while disinformation operations use accounts, messages, and coordinated activity to shape what audiences see and believe. Evidence must therefore show a change in political activity, not merely a technical outage.

The article reports that antibot4navalny found bot traffic “steady as ever” after the strikes. On the same Monday that Zelenskyy claimed disinformation had declined, several Kremlin-aligned accounts posted videos falsely saying experts wanted the European Union to lift sanctions on Russia. These continued posts directly challenge the idea of a political shutdown.

Danish analyst Valdemar Osted also said interactions on MAGA and AfD accounts had increased since the strikes. The article concludes that Yandex’s web crawlers appear to have been affected, while political disinformation continued. It also says charts showing lower bot activity appeared to be AI-generated.

The reported targets were not limited to one country or one political movement. The article mentions MAGA-affiliated accounts in the United States and accounts associated with Germany’s far-right Alternative for Germany, or AfD, party. It also describes broader Kremlin-aligned activity that promoted false claims about European Union sanctions on Russia.

The clearest example came on Monday, when several Kremlin-aligned accounts posted videos falsely claiming that experts were demanding an end to the EU’s ongoing sanctions against Russia. The article also cites Danish analyst Valdemar Osted, who said interactions on MAGA and AfD accounts had increased since the strikes. These examples suggest continued attention to politically polarized audiences.

The article does not identify every account, operator, or audience involved. It does say antibot4navalny reported steady bot traffic, despite claims that activity had fallen. It also notes that some charts showing reduced bot activity appeared AI-generated, weakening the case for a genuine decline.

Researchers distinguish genuine engagement from coordinated activity by examining patterns across many accounts. Useful signals include repeated wording, synchronized posting, unusually rapid sharing, similar account histories, and dense links between accounts. No single clue proves automation. Stronger conclusions come from several matching signals and from comparing activity over time.

They can also compare engagement before and after an event, inspect whether accounts behave like ordinary users, and check whether apparent charts or datasets are authentic. In this case, antibot4navalny tracked bot traffic, while Valdemar Osted compared interactions on MAGA and AfD accounts after the strikes. Those checks produced results that differed from public claims of a decline.

The article does not describe researchers’ full methods, so these are established ways to investigate such activity. Its central warning is clear: lower visibility or personal impressions do not prove that disinformation has stopped. Independent tracking found steady bot traffic, while some supposed decline charts appeared AI-generated.

A claim about reduced disinformation becomes disinformation when it presents false or unsupported information as fact. It can change how people interpret later posts, lower their skepticism, and hide ongoing manipulation. The article calls the claims about a post-strike decline ironic because the alleged evidence did not show that Russian political activity had stopped.

The article gives several examples. Zelenskyy said there was less hate, division, and provocation, while users claimed lower traffic around MAGA and AfD accounts. Yet antibot4navalny reported steady bot traffic, and Kremlin-aligned accounts continued posting false sanctions claims. The article also says charts showing reduced bot activity appeared AI-generated.

Social-media algorithms can help such claims spread when they reward attention, emotional reactions, or rapid interaction. A dramatic “disinformation is gone” message may attract shares before it is checked. The supplied article does not explain algorithm design in detail, but its evidence shows why independent verification matters before repeating reassuring claims.