Table 1: Organizational Form and Prevalence of Social Media Manipulation in Brazil
Initial Government
Politicians & Private
Civil
Society Citizens
&
Report Agencies
Parties
Contractors
Organizations Influencers
2010
Evidence found
Evidence
Evidence found
Paid supporters
found
(Eg. AM4
Virtual Activist
(Eg. Social
Company,
Movement (MAV)
Liberal Party, Havan, Yacows,
Businessmen (Eg.
Eduardo
Steve Bannon)
Paulo Maurinho)
Bolsonaro,
Social media,
Flávio
bloggers, journalists,
Bolsonaro,
and
Taíse de
influencers (Jouberth
Almeida
Souza, Allan dos
Feijó,
Santos, Oswaldo
Eduardo
Eustáquilo)
Guimarãe)
Source: Authors’ evaluations based on data collected. Blank spaces indicate no evidence was found.
Strategies, Tools, and Techniques
Whilst computational propaganda has been in use in Brazil since 2010, it was particularly
evident in the 2018 elections, and “fake news” became a major concern. Authorities were
handling sanctions quickly, and the Superior Electoral Court (Tribunal Superior Eleitoral, TSE)
subpoenaed Facebook to remove over 196 pages that contained false information, including
the page of Movimento Brasil Livre, a major force in the impeachment of Rousseff and
Bolsonaro’s rise (Maleronka & Declercq, 2018).
Brazil has shown quite an innovative spirit towards political propaganda and the 2018 election
was marked by a few incidents illustrating this characteristic. First, the concept of politicians
as social media celebrities and influencers is not a novel phenomenon in Brazil. In fact, the
current President Jair Bolsonaro started doing live-broadcasts and putting videos on YouTube
as early as 2016. Interestingly, this approach took on a different direction when certain
politicians became carriers of disinformation themselves. Congresswoman Joice Hasselmann
spread false information accusing the Rousseff’s Workers’ Party of having ties with the
Hezbollah (Joao Filho & Felizardo, 2018). She also alleged that one of the mainstream media
companies had a contract worth up to 600 million reais (approximately £130 million) to support
the Workers’ Party.
In the weeks leading up to the first round of elections, which took place on 3 October 2018,
Brazilians complained about the large volume of disinformation circulating on social media.
However, research by the Oxford Internet Institute has shown that traffic on Twitter was
particularly low (Machado et al., n.d.). Multiple reports released in the period between the first
and second rounds of voting, along with findings from investigative journalists, showed that
the bulk of Brazilian disinformation was being disseminated through WhatsApp. Over 120
million Brazilians use the platform regularly, making it an effective way of reaching the entire
population (Paulo Higa, 2018). As research has shown, the total volume of disinformation
spread over WhatsApp was at least eleven times greater than the total dissemination of junk
news on Twitter, and clustered networks of WhatsApp groups were used to disseminate
disinformation to thousands of users (Machado et al., 2019).
60