AfD Dominated TikTok In German Election

University of Copenhagen

The right-wing party Alternative für Deutschland (AfD) dominated TikTok during the German federal election in 2025. Its visibility was primarily driven by regular users - not by the party itself.

Photo: Solen Feyissa / Unsplash
Photo: Solen Feyissa / Unsplash

Social media is playing an increasingly important role in political election campaigns. TikTok and Instagram have become battlegrounds where politicians compete for the attention of voters, particularly younger ones.

In the most recent Danish general election, we saw a significant increase in the use of TikTok. This platform was also widely used during the German federal election in 2025. Researchers from the University of Copenhagen have examined how the campaign unfolded in the digital sphere.

AfD dominates

The researchers behind the study analysed nearly 48,000 TikTok videos from February 2025. They measured how often different parties were mentioned in hashtags (#).

The result was clear: if a winner were to be crowned based solely on TikTok visibility, it would have been a landslide victory for the right-wing party Alternative für Deutschland. The party topped the hashtag rankings every single day during the campaign.

On election day, the hashtag #afd was used almost three times as often as the hashtag of the second most-mentioned party.

"What is remarkable is that AfD's visibility does not primarily come from the party's own politicians, but from ordinary users who either support or criticise the party," says Jochen Kinast, postdoc and visiting researcher at the Department of Communication.

It is important to note that this study focuses solely on visibility on TikTok, not on support for party policies. Hashtags can be used to signal support, express criticism, or appear in politically neutral statements.

"Although the analysis focuses on Germany, the mechanisms are the same in Denmark and other countries. As more young people form their political impressions via TikTok, it becomes crucial to understand and discuss how the platform's algorithms influence what they see," Kinast adds.

Algorithms reward conflict

According to the researchers, the results suggest that TikTok's algorithms may favour polarising and conflict-driven content, regardless of political position. This means that parties that evoke strong emotions may come to dominate the political agenda on the platform.

"When both positive and negative attention increases visibility, it can amplify the perceived importance and size of a party, even if this does not reflect actual election results," says Assistant Professor Franziska Marquart.

The study shows that other parties could not match AfD on TikTok. For example, the governing party Die Grünen, which secured 39 more seats than AfD in the German parliament, was almost absent on TikTok during the campaign. Instead, it focused more on Instagram.

"TIkTok rewards attention and conflict - not necessarily political representativeness. Our research highlights that election campaigns today largely unfold on commercial platforms with their own logics. This calls for regulation of these platforms and greater transparency if we want to ensure that digital election campaigns do not distort democratic debate," says Marquart.

About the study

  • The study '#bundestagswahl: Hashtags and party visibility in the 2025 German federal election on TikTok' was published in the journal Communications in May 2026
  • The analysis is based on approximately 48,000 TikTok videos from the February 2025 election campaign
  • AfD was the most visible party in hashtags every day throughout the period
  • Visibility was primarily driven by ordinary users, not the party's own accounts
  • In the Bundestag, AfD is the sixth-largest party with 79 out of 630 seats

/Public Release. This material from the originating organization/author(s) might be of the point-in-time nature, and edited for clarity, style and length. Mirage.News does not take institutional positions or sides, and all views, positions, and conclusions expressed herein are solely those of the author(s).View in full here.