During a Diwali celebration event organized by the Bharatiya Janata Party (BJP) in Delhi, PM Modi stated, "Deepfake poses one of the biggest threats to India, potentially giving rise to misinformation and chaos."
He urged the media to educate the public about artificial intelligence and deepfake technology.
It is worth noting that a deepfake video featuring PM Modi surfaced recently, falsely depicting him participating in a Garba dance.
The Prime Minister emphasized the need for awareness and vigilance to counter the misuse of deepfake technology, which has the potential to create disarray and undermine democratic values.
PM Modi's comments come in the wake of the recent viral deepfake video featuring actress Rashmika Mandanna, causing widespread controversy on social media.
Notably, Bollywood icon Amitabh Bachchan had urged action against the misuse of deepfake technology through a tweet.
Simultaneously, a deepfake image featuring Sara Tendulkar, the daughter of former Indian cricketer Sachin Tendulkar, surfaced.
The manipulated photo falsely suggested that she had taken a picture alongside Indian cricketer Shubman Gill. However, the reality was that in the original photo, Sara Tendulkar was standing with her brother Arjun.
The deepfake technology had ingeniously substituted Arjun's face with that of Shubman Gill.
Adding to the concerns, a deepfake video of Bollywood actress Kajol is currently circulating widely.
The video depicts the actress in a lewd manner, manipulated through deepfake technology. Within the edited footage, the actress is shown undergoing a wardrobe change.
This explicit deepfake video of Kajol is gaining widespread attention on various social media platforms.
In a recent development, the Ministry of Electronics and Information Technology (MeitY) has issued an advisory to prominent social media platforms such as Facebook, Instagram, and YouTube.
The advisory instructs these platforms to promptly remove any deepfake content within a 24-hour timeframe. The government has explicitly stated that failure to comply with this directive will result in criminal and judicial actions being taken under Indian laws.
The term 'deepfake' is derived from 'deep learning,' a subset of machine learning. Deepfake technology is based on artificial neural networks and employs multiple layers (hence, 'deep').
It creates content that appears authentic, even though it may not be. The first notable instance of deepfake technology emerged in 2017 when a Reddit user created numerous deepfake videos.
The creation of deepfake videos involves two networks: an encoder and a decoder. The encoder reads the original content and transfers it to the decoder network, which generates the fake video with a different face.
This process results in a video where the face has been replaced, while the rest of the video remains the same.