As Brazil picked up the pieces on Monday from the damage caused by supporters of ex-President Jair Bolsonaro, it was hard not to see echoes of the Jan. 6, 2021, insurrection at the US Capitol — not least because both incidents were stoked by activity on social media. Brazilian users of Meta Platforms Inc.’s WhatsApp, as well as Telegram and TikTok, saw a surge in calls to attend a “war cry party” in the capital in the run-up to Sunday’s violence in Brasilia, according to the Washington Post.
Even before that, ahead of the presidential election last October, Twitter, Facebook and WhatsApp had been flooded with misinformation about the integrity of Brazil’s vote, which saw Luiz Inácio Lula da Silva narrowly defeat Bolsonaro. It doesn’t help that Meta allows misinformation from political candidates to go unchecked, a policy it should have changed long ago.
But it’s not just Meta.
A search of #bolsonaro on TikTok on Monday morning led to at least two popular videos praising rioters as “patriots” or fighting for “freedom.” YouTube channels touting Bolsonaro’s claims of election fraud garnered tens of millions of views ahead of the vote. And Brazilian election deniers have seen a surge in followers on Twitter, according to an analysis by Rest Of World, a nonprofit journalism organization.
It all highlights the bigger problem: Social media firms are still not investing enough on controlling misinformation that can spin out of control, relying on cheaper methods such as software and contractors to weed out harmful content and — critically — underfunding such efforts outside the US. The whistleblower Francis Haugen, who exposed a range of harms neglected by Facebook in 2021, zeroed in on the company’s inadequate efforts to tackle misinformation in foreign countries like India. She warned about forthcoming global elections and said the site “cuts corners and provides unequal and inadequate defenses.” In the case of Twitter, new owner Elon Musk reportedly fired the company’s Brazilian staff after he took over.
Ironically, Facebook and Twitter made impressive strides in preventing large misinformation campaigns from disrupting US midterm elections last year, according to multiple surveys. But those platforms have a long way to go elsewhere.
“Sometimes we see content staying up in Spanish and Portuguese far more than in English, even when its English counterpart has been taken down,” says Roberta Braga, a Brazilian-American researcher with Equis Research, a public-opinion research group focused on Latinos.
“They don’t focus anywhere like they do in the US,” says Jiore Craig, who runs elections research for the Institute for Strategic Dialogue, a London-based think tank that tracks online misinformation. She says that online narratives leading up to the Brazil riots looked similar to those used in the run-up to Jan. 6, and have also popped up ahead of elections in France and Australia, where posters use language like “traitors” and “fraud” and English-language hashtags like #StopTheSteal.
One big difference with the US is that WhatsApp has become one of the most popular platforms for spreading misinformation about the election, according to Braga.
While more radical Brazilians will go to Telegram, a loosely moderated messaging and broadcasting app, to engage with likeminded groups, the vast majority use WhatsApp to communicate with friends and family and conduct business, where many election-denying narratives gained a mainstream foothold. Braga, whose own family and friends in Brazil have become radicalized Bolsonaro supporters, says WhatsApp could do more to mitigate the reach of its broadcasting tools, which “exacerbate the circulation of misinformation, like it or not.”
WhatsApp has tried since 2018 to impose stricter limits on message forwarding to make it harder for misinformation about election integrity or Covid-19 vaccines to go viral. It also adds a label saying “forwarded many times” to help slow down the spread of rumors and fake news. The company should do more, but that would mean upsetting advocates of free speech and privacy: much of the incendiary content is being spread in private groups which WhatsApp cannot see or interfere in, since all communications are encrypted.
Complicating matters is that viral content on WhatsApp or Telegram often starts somewhere else, like YouTube or TikTok. “Much of the activity is cross platform,” says Craig. “One place you might see the narrative and the other you amplify the narrative. Taking down one doesn’t take down the other.”
To tackle the problem, social media firms will need to step up efforts to communicate with one another about forthcoming elections and election-fraud claims. Researchers also say social media firms need to be more transparent about how content flows and gains traction on their sites, to help them find new ways of stopping it from spiraling out of control.
But a big part of the problem is also, admittedly, out of their hands. Bolsonaro — like Donald Trump — laid much of the groundwork for the stolen-election narrative himself, having spent months claiming unfair treatment at the hands of Brazil’s election authority.
“What we are seeing in Brazil is very similar to what we saw in the US,” says Braga: “the trajectory of how the election-fraud narrative penetrated society, and the types of influencers spreading it.”
More From Bloomberg Opinion:
• Musk Drags Twitter Down a Dangerous Rabbit Hole: Parmy Olson
• Jan. 6 Panel Makes Trump Prosecution Imperative: Timothy O’Brien
• How Peaceful Protest Lost Its Power: Leonid Bershidsky
This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.
Parmy Olson is a Bloomberg Opinion columnist covering technology. A former reporter for the Wall Street Journal and Forbes, she is author of “We Are Anonymous.”
More stories like this are available on bloomberg.com/opinion
©2023 Bloomberg L.P.