Going Viral

A Correction, Not a Deepfake

Celebrities and influencers have become entangled in public health discourse, and are sometimes given more authority than scientists or public health officials. And their reach, like the misinformation they spread, is amplified by the same neural-network recommendation algorithms that power the platforms they are on.

Generation process

Unlike today’s diffusion models, which generate images by gradually denoising random noise over dozens of steps, the videos in Going Viral are made using a Pix2Pix conditional generative adversarial network (cGAN). In a cGAN, a neural network is trained on sets of two images where one image becomes a map to produce a second image. In Going Viral, the two images are landmarks from facial recognition and a frame from a video. Next, we take video of an expert or journalist speaking on a topic the influencer has spread misinformation about and extract the facial landmarks from that expert. We then use the facial landmarks of the second speaker to generate video frames of the influencer, celebrity, or politician speaking the same words. Finally, the new frames are combined with the audio track of the expert or journalist to produce a new video where an influencer is correcting the misinformation they have spread.

Currently twenty-one politicians, influencers, and celebrities from nine countries are part of the project, with new figures added as new misinformation campaigns emerge.