The Global Risk Report 2024 of the World Economic Forum has ranked India as the top country where the crisis of misinformation and disinformation is most severe.
The world’s largest democracy India is gearing up for its parliamentary elections this year. However, amidst this preparation, India faces a significant challenge of dealing with AI-generated misinformation, which can deepen social and political divides by targeting people’s existing beliefs.
Relying on news sources, more than half of internet users in India consume online news. However, with a staggering 1000% increase in AI-generated misinformation, distinguishing between real and fake news has become difficult for citizens.
Neuroscientist and author Mukti Kulkarni highlighted in an academic speech that the problem of AI-generated fake news is now spreading to politics. For instance, he mentioned a fake video of the popular television show “Kaun Banega Crorepati” being used to boost a political agenda.
Additionally, AI-generated fake videos have been observed showing deceased leaders like Swami Vivekananda delivering speeches. Kulkarni also pointed out how several propaganda websites use AI to promote a narrative similar to that of the Chinese government.
Nature of Misinformation in India
Karen Rebelo, deputy editor of fact-checking agency BOOM, observed that most fake news that tend to go viral in India are mostly politically or religiously motivated.
“However, there has been a blurred line between the two recently, especially after the inauguration of the Ram Temple,” she said.
The ruling Bharatiya Janata Party’s (BJP’s) political core is embedded in the Hindu nationalist agenda that asserts the primacy of Hindu values in Indian society. However, this contradicts India’s secular principles, leading to deepening religious divides, as reported by the media.
This religious polarisation exaggerates the impact of misinformation, said Joyojeet Pal, assistant professor at the University of Michigan.
Specialising in the misinformation landscape in India, he added that it is because people are interested in negative narratives about opposing communities.
This was seen during the 2020 coronavirus pandemic when fake news surged. False allegations were made against Indian Muslims, accusing them of conspiring to spread the virus. This misinformation led to Muslims facing daily threats.
In the later half of the year, the hashtag #NoMeat_NoCoronaVirus was trending on X (formerly Twitter). Users were ‘appealing’ to others to quit meat consumption to safeguard everyone from Covid-19.
Psychologist Dr. Sheikh Abdul Basir explained that repeatedly exposing people to something can mentally prepare them to believe in it. In this way, AI-generated fake news can target people’s existing beliefs, stereotypes and prejudices.
Role of Generative Artificial Intelligence in Elections
So far, audio and video content has been trimmed and shared in a certain way. However, with the advent of generative artificial intelligence, fake audio, video, images, and text can be technically created within seconds. This technology simplifies deceptive political propaganda.
While the use of generative AI has been ongoing for years, its use in elections was seen in 2020 when Manoj Tiwari, a BJP candidate from Delhi, used AI technology to reach out to a pool of voters in three languages – Hindi, English and Haryanvi – two of which were dubbed using deepfake.
Neelkanth Bakshi, BJP’s co-incharge of social media and IT in Delhi, told VICE in 2020 that Deepfake technology has significantly aided Delhi BJP’s campaign efforts.
Through deepfake technology, faces of individuals in videos are replaced so seamlessly that it becomes difficult to distinguish them. Such usage of AI-generated fake videos was also observed in state assembly elections last year.
The regional political party of Telangana Bharat Rashtriya Samiti (BRS) wrote a letter to the Election Commission of India (ECI) accusing the opposition Congress party of using deepfake technology to influence voters.
In one of the deepfake videos circulated online, BRS leader and Telangana minister Chamakura Malla Reddy were seen dancing and wooing voters by saying, “You will get jobs if you vote for [K. Chandrashekar Rao] KCR.”
Moverover, BOOM found in its investigation that AI-generated voice clones of Shivraj Singh Chauhan and Kamal Nath, candidates from two national parties BJP and Congress, were widely circulated in Madhya Pradesh.
Uzair Rizvi, expert in digital verification, said that most AI-generated fake news originates from the IT cells of political parties. “These IT cells are making India’s misinformation landscape more complex,” he said.
This also increases the possibility of “liar’s dividend,” meaning that even legitimate media content can be questioned to be fake in a politician’s favour.
How to Combat Fake News
Several fact-checking agencies attempt to debunk online misinformation daily. However, their work faces many challenges, such as the high cost of tools for identifying AI-generated content compared to those for creating them.
Rebelo said, “The simple reason is that there’s not much money to be made in detection.”
As a result, the number of people producing AI-generated content surpasses the number of fact-checkers, hindering efforts.
Union Minister of Communications, Electronics, and Information Technology Ashwini Vaishnav told ANI that they are working against Deepfake through four pillars – identification, prevention, strengthening the system, and awareness.
He also mentioned that efforts are being made to introduce new laws to combat deepfake.
Currently, specific laws address tech misuse in India, including Copyright Violation and Defamation. In addition, IT Rules, 2021 mandate takedown of reported fake video, especially a deepfake, within 24 hours of reporting.
Due to Centre’s pressure on social media giants against misinformation and Deepfake videos, Meta, and other social media companies, are under heavy pressure to identify and remove harmful AI-generated content on their platforms.
Meanwhile, Pal commented that despite assurances from social media giants regarding trust and security standards, “they have reduced their investment in this area.”
In such a scenario, it becomes a responsibility on netizens as they must verify whatever they read, watch, or listen online, especially content that uses emotionally charged language.
Yet, many, like first-time voters Kashish Khan and Tushar Raghav, neglect fact-checking due to busy lifestyles or prefer reading newspapers over social media.
A digital news consumption survey conducted by researcher Tanushree Basuroy in March 2023 stated that 43 percent respondents believe that observing how news spreads and its absence from other digital platforms was a common method they used to spot online misinformation.
India could also benefit from adapting a similar systemic media literacy as in Taiwan. Melody Hsieh’s organisation in Taiwan, Fake News Cleaner, conducts anti-disinformation workshops in various settings in Taiwan including universities, temples, and fishing villages, occasionally offering incentives like handmade soap to encourage participation.
Tasneem Zahra is a journalist and documentary filmmaker based in South Asia. Her work has appeared in Deutsche Welle, The Wire, The Better India, The Citizen, and more.
