A new report from Freedom House, a D.C.-based nonprofit political advocacy group, found that at least 16 countries over the last year have used deepfake technology to “sow doubt, smear opponents, or influence public debate.”

Earlier this year, during national protests in Pakistan, Former Prime Minister Imran Khan (who is now imprisoned) posted a video on X to rally his supporters. That video used deepfake technology to show an image of a protestor standing in front of riot police - this image was generated by AI.

Former President Donald Trump and Florida Governor Ron DeSantis have both used deepfake technology to attack each other in political ads as the Republican primary competition has heated up this year. Trump ran a fake Twitter Spaces event that showed DeSantis in a conversation with George Soros, Adolph Hitler, and…. The Devil. The DeSantis team fought back using AI to create fake images of Trump hugging and kissing Dr. Anthony Fauci.

China has been hiring private companies to deploy AI-generated propaganda that mimics newscasters, they even created an AI-generated news division called “Wolf News” that shares video clips of human characters presented as newscasters. Venezuela has employed similar tactics using U.S.-based AI avatar company Synthesia.

All of this has an important consequence: people no longer know what is real and what is deepfaked. The lines of reality are becoming blurred. This allows political actors to call out video and audio recordings as deepfakes, even if they are genuine, further spreading doubt over all content. The technology is leading us to a world where you can no longer trust what you see, read, or hear on the internet as genuine content - even if it looks or sounds like a real well-known political figure.

Aside from using AI to create false narratives, it was found that at least 22 countries had processes in place for social media companies to use AI to identify and remove dissident political speech from the internet. The report found that a record 41 governments, including the U.S. and European countries, blocked websites with content that should be protected under free expression standards within international human rights law.

Deepfakes will only get better as AI technology advances. Without proper laws and regulations, it will become harder to decipher what is real and what is fake on the internet. How we handle this issue will have lasting implications for humans everywhere.

Originally featured in The AI China Report #3