The lie arrives before the vote
Disinformation is no longer a sideshow to an election. It is increasingly part of the campaign itself: industrial, automated and designed to travel faster than verification. Its purpose is not always to persuade voters of one particular truth. Often, it is to make truth appear impossible to establish.
Bangladesh offers a stark example. On the eve of the election, Facebook was flooded with fabricated quotations, doctored photographs, misleading graphics and forged documents. A review of 220 cases recorded more than two million engagements within the first 24 hours. The material included 104 videos, 44 photocards, 75 items involving forged documents or fabricated quotes, and 58 using deepfakes or other artificial-intelligence-generated content.
The figures expose the central advantage enjoyed by disinformation: speed. A false claim can be produced in minutes, copied thousands of times and absorbed by millions before a credible correction has even been prepared. By then, the correction is competing not only with the original lie but also with the suspicion that every source is manipulating the public.
Deepfakes become political weapons
The United States midterm elections have demonstrated how sophisticated this machinery can become. Researchers attributed a campaign known as Matryoshka, or Operation Overload, to Kremlin-linked actors. The operation circulated manipulated videos purporting to show celebrities, including Alyson Hannigan and Eric Roberts, attacking Democrats.
These were not crude fabrications easily dismissed by a discerning viewer. The videos used stolen footage, counterfeit CNN branding and coordinated bot activity to create the appearance of legitimate reporting and spontaneous popularity. Artificial amplification is crucial: viewers are more likely to trust a claim when it appears to be spreading widely, even if that apparent popularity has been manufactured.
This is the psychological architecture of modern disinformation. The fake video supplies emotional shock; the false branding supplies authority; the bots supply social proof. Together, they create a counterfeit public reality.
Scale turns falsehood into atmosphere
NewsGuard reported on October 26 that it had identified 32 false claims about the 2026 midterms. Those claims appeared 432 times across 12 platforms, 256 social-media accounts, 64 websites and 24 national cable-news segments. X accounted for 57 per cent of confirmed instances, while 14 claims alleged election fraud.[1]
The danger lies not only in individual lies but in their accumulation. Repeated allegations of stolen elections, illegal ballots or hidden conspiracies can weaken public confidence even when each claim is debunked. A voter need not believe every falsehood for the campaign to succeed. It is enough to leave the impression that no result can be trusted.
That impression is particularly dangerous after polls close. Democratic systems depend on losers accepting legitimate outcomes and citizens believing that institutions can resolve disputes. Disinformation attacks both conditions at once.
The local-news disguise
The manipulation does not stop at social media. NewsGuard also identified 22 partisan websites launched in 2026 while presenting themselves as independent local-news outlets, bringing the reported total to 1,181 such sites.[1] Their strategy is more subtle than a deepfake. They imitate the visual language of journalism, publish selected facts and insert partisan narratives into apparently familiar community coverage.
This “pink slime” model exploits a valuable public instinct: people tend to trust information that appears local. A fabricated national story may be viewed with caution; the same claim, presented as neighbourhood reporting, can seem more credible. The result is an information environment in which propaganda borrows the costume of public service.
Rumour crosses every border
In Ukraine, anonymous Russian Telegram channels falsely claimed that half of the country’s State Emergency Service personnel would be sent to the front. Ukrainian fact-checkers identified the allegation as fabricated. The rumour was calculated to provoke fear, confusion and anger—especially among communities already living under the pressure of war.
This illustrates why disinformation should be understood as a security threat, not merely a problem of online accuracy. False reports can divert emergency resources, endanger public servants, inflame social divisions and undermine confidence in institutions responsible for public safety.
The democratic response
The answer cannot be censorship by default, nor can it be the expectation that citizens independently investigate every video, graphic and headline they encounter. Platforms must disclose coordinated manipulation, label synthetic media clearly and preserve evidence needed by researchers. Election authorities and newsrooms must publish rapid, accessible corrections. Schools must teach verification as a civic skill, not an optional digital habit.
Most importantly, democratic institutions must communicate before rumours fill the vacuum. Silence is easily colonised by conspiracy.
Disinformation succeeds when speed overwhelms scrutiny and distrust outlasts evidence. Protecting elections therefore requires more than defending ballot boxes. It requires defending the public’s ability to distinguish reporting from theatre, disagreement from deception and a genuine political choice from a reality manufactured for them.