Deepfakes and AI used to spread political disinformation in various elections

Research and investigations highlight the increasing use of deepfakes and generative artificial intelligence as tools for political disinformation during major electoral processes. Studies analyzing presidential elections in Ecuador, the United States, and Mexico examine how synthetic media can realistically imitate the voices and images of public figures, making detection difficult and undermining content credibility.

In a specific case involving Colombian politics, digital clones of Senator Iván Cepeda were used to disseminate false claims via social media prior to a presidential runoff. These deepfake videos utilized AI to clone his voice, attributing fabricated statements regarding expropriations, militia incursions, and other sensitive topics to him. The disinformation campaign, which garnered over 700,000 views, targeted specific voter niches through paid advertising on Facebook that bypassed Meta’s transparency mechanisms.