Deepfakes enter Indian electionDate: 22 February 2020 Tags: IT, Mobile & Computers
A day before Delhi elections, two videos of a political party president of Delhi unit urging citizens to vote for the party, in english and haryanvi, were sent to 15 million voters via 5800 WhatsApp groups. The videos were deepfakes, digital media firm Vice reported.
Advances in Artificial Intelligence (AI) and Machine Learning (ML) have enabled computer systems to create synthetic videos, a.k.a. deepfakes.
A deepfake video can show a person saying or doing something that they never said or did. An AI algorithm is taught, using deep learning, to manipulate actual video and audio to mislead viewers.
Deepfake online videos are burgeoning as tools and applications to make fictional content gets cheaper, lowering barriers for non-experts to make them.
One of the ways to make deepfake videos involve encoding, decoding, and face-swapping. First, an AI encoder runs several face shots of two persons.
Then, the encoder learns of similarities in the faces and compresses the images.
As the faces are different, one AI decoder picks the first person’s face, and another decoder picks the second person’s face.
Then, the encoded images are fed into the "wrong" decoder to perform the face swap. This process is done on every frame to make a convincing video.
In certain other cases, the AI is trained to create new images and videos from scratch using Generative Adversarial Network (GAN).
GAN, with the help of Machine Learning, makes two neural networks to contest against each other in a game with a given training set. This helps generate new data and output related to the training set.
Previously, a deepfake video by Future Advocacy showed UK's Prime Minister Boris Johnson and Labour leader Jeremy Corbyn endorsing each other. The think-tank's video was posted in an attempt to show the potential of deepfakes in undermining democracy.