Indian Politician Manoj Tiwari Used Deepfake To Reach More Voters

Deepfakes and misinformation are being eradicated from the social media platforms due to the upcoming US elections. Amidst all this, an Indian politician has used artificial intelligence to make deepfake videos, making it the first use of AI deepfake in India’s political scenario. In the deepfake video, the politician is seen saying things he didn’t intend to say in the original video.

As reported by Vice, BJP politician Manoj Tiwari, who was contesting the Delhi CM elections recently, speaks in English in the original version of the video and in the manipulated video, he uses a Hindi dialect of Haryanvi.

Below is the original version of the video in question:

And, here is the deepfake video version:

The Ideaz Factory, a political communications firm stated that it worked with Bhartiya Janata Party for their social media campaign. Their main focus was to create positive campaigns using deepfake technology to reach to masses speaking different languages.

The firm said it used a lip-sync deepfake algorithm, trained with Tiwari’s speeches to convert audio sound into basic mouth shapes. An official said that this step allowed the candidate to reach voters he was unable to connect with due to the linguistic barrier. As per the stats, the deepfake video reached around 15 million people all over in India.

Previously, an infamous deepfake video of former US President Barack Obama raised the question that how misleading videos can be used as a weapon to spread false information at the time of elections.

In October 2019, a bill was passed in California against the spread of deepfake videos. It said that sharing a deepfake video of a politician within 60 days of an election will be considered illegal.

Several social media organizations have declared plans to tackle deepfakes on their respective platforms. Reddit has updated its policy to ban all impersonation on the platform that also includes deepfake. Facebook has started banning deepfakes since the last month. Twitter is also going to launch the deceptive media ban from March 2019.