New research exposes AI bot scheme spreading pro-Trump content, attacking Biden-Harris-Walz on X

The Clemson University report sheds light on an ongoing social media influence campaign using AI-powered bots to sway political discussions in U.S. federal elections.

According to the findings, large language models are behind the creation of content that appears organic, strategically placed in the replies of real users' posts.

The network identified consists of at least 686 accounts, collectively responsible for over 130,000 posts since it began in March this year.

Research found that the campaign targets both Republican candidates during primaries and Democrats in general elections while also advocating for specific political issues.

Some examples of their posts are:


The accounts in the network all feature European-sounding first and last names, with only a few providing biographical details. Those that do often signal conservative values, as per the research, with phrases like "Girl Mom :)" or "Christ is King!"

A small fraction of these accounts even display a blue check mark, indicating they are X premium subscribers.

The report points out that the network's primary goal is to influence perceptions of U.S. political candidates, with a particular focus on the 2024 presidential election.

The campaign shows clear support for former President Donald Trump while frequently attacking the Biden-Harris-Walz team.


Although activity was initially low, it surged around the time of the Biden-Trump debate in late June, with some days seeing over 1,000 replies promoting Trump or criticising his opponents.

To stay updated on the stories that are going viral, follow Indiatimes Trending.