New research from Graph has found that Chinese state-linked operations have become more aggressive in their efforts to influence the 2024 US election. Campaigns have been seen using fake profiles of American voters on social media platforms to discuss sensitive social issues and talking points that spread divisive rhetoric ahead of the presidential election.
Influence operations (IOs) were found to have used AI-generated images of US voters, featuring realistic avatars that were likely produced using a commercial AI video creation platform created in the UK.
Despite their compelling content, the campaigns were mostly categorised as “spamoflauge”, as the videos were of low quality and “spammy” in nature. None of the analysed videos received more than 300 views and generated very little authentic engagement, highlighting the difficulties in producing compelling political content.
Fake news is old news
The report shows evidence of “coordinated amplification,” whereby fake accounts shared the same content and posts across a disinformation network. The content typically sought to undermine democracy and the U.S. political process, and promoted debate on sensitive topics such as the legitimacy of the 2020 election and anti-establishment messaging.
This is not the first report to establish that China has been using (or at least supporting) campaigns to influence American citizens. The state appears to be less interested in backing a specific candidate and more focused on dividing the American public and create distrust in the American political system.
The use of AI in political influence campaigns is here to stay, and as the technology evolves, it will become harder to detect. Most AI content creation platforms claim that their services are not intended for political use, but moderating its application is difficult.
