KOCHI, India — When Prime Minister Narendra Modi and his right-wing Bharatiya Janata Party swept to power five years ago, the spread of online misinformation wasn’t on the radar of most Indians.

But the BJP’s successful use of misinformation to promote its Hindu nationalist agenda has demonstrated how easily voters are swayed by what they see online.

“People are spreading misinformation unknowingly. That is a tragic thing,” said K. Anvar Sadath, the executive director of Kerala Infrastructure Technology and Education (KITE), which works with 2,000 school campuses across the state to provide digital literacy programs.

Just in the southwestern state of Kerala, recent false posts have included a barrage of claims around the Sabarimala temple, a site of controversy over women’s access, where false posts included: a devotee committing suicide after female protesters entered the temple, police arresting a child protester at the temple, and a BJP leader claiming a woman was injured due to police brutality there.

During this election, which kicks off Thursday, social media platforms, fact-checking publications and local educational organizations like Sadath’s are working furiously to combat bad information. With an estimated 900 million Indian citizens eligible to vote, it has become a priority to ensure that voters aren’t being deceived by what’s sometimes called “fake news.”

“Fake news in the Indian context isn’t necessarily news that’s incorrect or false. It’s more like a manipulation of facts or propaganda,” explained Maya Mirchandani, senior fellow at the Observer Research Foundation, which last year published a study called “Digital hatred, real violence: Majoritarian radicalization and social media in India.”

Mirchandani warns that what she calls “signalers,” who are savvy enough to mask political propaganda as political opinion, are especially influential.

“If I say, ‘You are a Muslim and you should be killed,’ I’m violating the code of conduct. But if I’m giving you a history of Muslim invasion and saying it’s time for us to stand up and claim our rights, I’m saying the same thing, but I’m not saying it in the language that’s going to be flagged,” she said.

Using this strategy, BJP signalers pushed out alarmist messaging that provoked religious violence from Hindu vigilantes all throughout the country. The party’s digital campaign team would also often fan the fires of its supporters by resharing videos or photos that would falsely demonize minority groups without proper context.

The Indian site Fact Checker found that hate crimes against religious minorities have spiked under BJP leadership, and that “Muslims, who comprise 14% of India’s population, were the victims in 62% of cases” over the last 10 years.

“If something happens in 2009, it’s one incident. Then you join it with something in 2012 and 2016 — now you have a narrative that makes people look bad, but you’re eliminating context,” Mirchandani said.

Efforts to create guardrails that regulate misinformation in India have proved difficult for several reasons.

First, fact-checking on platforms like Facebook are often done in English or Hindi. With over 120 languages spoken in India, it has been hard to regulate misinformation in local languages.

The popularity of WhatsApp, an encrypted chat platform that allows users to speak freely without regulation in closed groups, also creates problems. While open platforms like Facebook and Twitter can crack down on shared misinformation, what’s shared via WhatsApp can only be seen by members of the closed group.

And although the Election Commission has developed guidelines to restrict political parties from posting misinformation in campaign efforts, this has not stopped supporters and influencers from doing so under the guise of political opinion.

Social media platforms have felt the pressure from India’s government to do something about the spread of misinformation as the election nears. In July 2018, WhatsApp rolled out a product change that removed the quick forward button and limited message forwarding to only five people (versus 250 people previously). But despite these efforts to prevent high volume distribution, there’s still a larger issue of digital literacy.

As the number of new internet users continue to rise in India, digital literacy is not rising at the same speed. The Kantar ICUBE 2018 Internet Usage Trend report predicts a 35 percent increase in internet usage by rural India and an 18 percent increase in all of India by the end of 2019. This means there’s a huge number of new Indian internet users who will be ill-equipped to question the authenticity of online content.

Prasanth Sugathan, the legal director of Software Freedom Law Center (SFLC), has been leading panel discussions on combating misinformation all throughout the country. He insists that while the government and intermediaries should create systems to monitor misinformation, helping people understand how to access correct information is most effective.

“There may not be much you can do with wrong information out there, but you can counter that with the right kind of information,” he said.

This article is part of a collaboration between The GroundTruth Project and the Knight Chair in Media and Religion at the University of Southern California’s Annenberg School for Communication and Journalism, made possible in part by the Henry Luce Foundation.