Politician Releases A Deep-Fake Video Of Himself To Reach More Voters

An Indian politician has harnessed the power of AI to create a deep-fake video of him saying things he did say. In the original version of the campaign video, the politician in question, Manoj Tiwari spoke English, but the fabricated version of the footage shows him spoke in Haryanvi.

The firm behind this AI-modified video is The Ideaz Factory, a political communications firm. It remarked that it was working with the Bharatiya Janata Party to make what it called “positive campaign” with the same tech that makes deep-fake videos possible. The voice that spoke Haryanvi was of an actor.

According to Sagar Vishnoi of The Ideaz Factory, the company makes use of a lip-sync algorithm that was trained with Manoj Tiwari’s real speech. After that, what the AI did was to create mouth shapes according to each audio sound.

The reason why the Bharatiya Janata Party thought of using a deep-fake video is to reach more voters in regions where English and Hindi are not spoken.

So far, the video has successfully reached over 15 million people in India.

The deep-fake algorithm is notoriously effective in creating nonconsensual pornography and misleading political speeches. A prime example is the fabricated video of Nancy Pelosi, a House Speaker of the U.S government.

The U.S has taken actions against the use of this tech. Back in October last year, California outlawed the sharing of deep-fake videos of politicians within 60 days of an election.

Large social media sites have also taken actions against the spreading of fabricated videos. Twitter, Facebook, and Reddit have already had their own policies to prevent misleading information being spread by deep-fake videos.

While the Bharatiya Janata Party and The Ideaz Factory said that the video of Manoj Tiwari speaking Haryanvi is for a positive purpose, who knows what will happen next.

>>> Mark Zuckerberg Appears In A Fake Video On Facebook-Owned Instagram