AI-Generated Disinformation Poses Growing Threat Ahead of Elections

AI as a New Weapon for Disinformation

With the upcoming elections in the United States, social media platforms are facing a flood of AI-generated fake content. Experts warn that the scale of disinformation has outpaced current moderation systems. The problem is exacerbated by leading technology companies reducing human moderation in favour of fully automated systems.

Today, AI tools make the creation of convincing photos, videos, and texts extremely cheap and accessible. Industry specialists note that modern technology allows for information campaigns on a scale that would have been considered science fiction just a few years ago. Disinformation actors often alter minor details in content to bypass automated filters designed to detect previously flagged material.

Real Examples and Moderation Flaws

A notable example of rapid disinformation spread involved a rumour regarding Texas Senate candidate James Talarico. On X, claims circulated that the candidate was being treated at a mental health centre, despite the fact he was actually resting at home with the flu. Although Talarico and the hospital quickly denied the report, the false narrative persisted for days and took on more severe forms, particularly within Spanish-language media spaces.

Studies show that platforms monitor Spanish-language content with less rigour than English content. When moderation relies solely on algorithms, language barriers and a lack of cultural context prevent systems from curbing the spread of falsehoods in time. Consequently, debunked rumours often return in new, more damaging versions.

What is the Stance of the Platforms?

Social networks employ varying approaches. Meta’s policy requires advertisers to disclose the use of AI, while YouTube prohibits disinformation related to election processes. However, many types of fake information remain outside these rules. In the case of X, the system relies primarily on user-generated "Community Notes."

The problem has moved beyond U.S. borders and is now global. In Spain, a political party faced legal action due to a deepfake video used during an election. The primary challenge is no longer just identifying synthetic media, but stopping false narratives that change instantly and spread across platforms before fact-checking can occur.

For Georgia, where social media serves as a primary tool for shaping public opinion, this trend should serve as a serious warning. In the future, convincing AI-generated disinformation may become even harder to detect, which could significantly undermine trust in electoral processes.