The recent advancements in AI and easy access to generative AI-powered tools have made it easier for anyone to create realistic deepfakes and share them online. This has led to a spike in social media posts containing deepfake photos and videos of celebrities and politicians.
A case in point is the American singer and songwriter Taylor Swift whose fake explicit photos were reportedly shared more than 45 million times on X last week.
Though X took down the posts, their delayed response allowed the deepfakes to proliferate. Later, X also blocked all searches related to Taylor Swift to prevent more users from discovering and sharing her deepfakes.
This has sparked a massive backlash against social media companies and X in particular for not doing enough to detect and prevent circulation of harmful content more quickly. Experts believe that X lacks the infrastructure to identify and block large scale spread of misinformation and deepfakes, especially after its acquisition by Elon Musk who dismissed half of the workers including 80% of the trust and safety team immediately after taking control in October 2022.
The recent increase in deepfakes targeting celebrities and politicians has also alarmed lawmakers across the world. In the US, White House Office called the Swift deepfakes “alarming” and pressed upon the need for a legislative action to protect people from fake AI generated images.
Though there are no specific laws to regulate AI deepfakes, in 2022, Biden administration had proposed a bill called Deepfake Task Force Act in the US senate to formulate a plan to curb the proliferation of digital content forgeries.
In India, social media companies were reminded in December 2023 by the Ministry of Electronics and Information Technology (MEITY) to comply with the IT rules and follow the directive on prohibited content. The ministry also asked social media intermediaries to submit a report on the action taken by them to deal with deepfakes.
The order came after women’s rights bodies expressed concern over the circulation of deepfake video of actress Rashmika Mandhana on social media platforms. In January, the Delhi police apprehended a 24 year- old from Andhra Pradesh for allegedly creating Mandhana’s deepfake viral video.
In addition to celebrities, several politicians including former US president Donald Trump and Indian prime minister Narendra Modi have been targeted by deepfake campaigns to spread disinformation.
In 2019, a deepfake video of Meta CEO Mark Zuckerberg was also posted on social media.
Deepfake is an artificially generated content from photos and videos of an actual person. To make the artificial content as realistic as the original, deepfake use sophisticated machine learning (ML) models such as generative adversarial networks (GAN), where a discriminator algorithm is used to learn and detect inconsistencies, while the generator algorithm tries to avoid detection creating new versions of the deepfake.
Image credit: Wikimedia Commons