By: Andre Nicola and Caroline Arkalji
Brazil’s 2026 presidential election exposed a basic electoral challenge: how can electoral authorities guarantee fair competition when private companies control the channels through which campaigns reach voters? As incumbent President Luiz Inácio Lula da Silva and Senator Flávio Bolsonaro faced off in Brazil’s October 25 runoff election, which Lula ultimately won, both campaigns accused U.S.-based Meta, owner of popular social media platforms Facebook, Whatsapp, and Instagram of treating them unequally across its platforms. Unclear advertising systems, inconsistent content moderation, and limited access to platform data can undermine electoral fairness even without partisan intent, creating a structural accountability problem. When private platforms control political advertising, content distribution, and the data surrounding their effects, electoral authorities may struggle to guarantee a level playing field. Even in a country where the electoral management body, the Supreme Electoral Court (Tribunal Superior Eleitoral or TSE), has sweeping powers compared to other countries, establishing enforceable standards remains a challenge.
In the case of Brazil’s recent elections, the controversies intensified during the latter half of the campaign period, as allegations emerged that Meta’s algorithms were limiting the reach of Lula’s posts, alongside reports of a temporary deactivation of his Instagram account. The Bolsonaro campaign also filed complaints with the TSE, alleging that algorithms allowed Lula’s campaign to evade oversight, including over posts accusing Bolsonaro of ordering Lula’s assassination.
Meta attributed these incidents to technical errors and gaps in its safeguards, denying deliberate political bias. The TSE ordered Meta to adopt 17 measures to ensure equal treatment of presidential campaigns following allegations of uneven content distribution, irregular political advertising, and inconsistent AI safeguards. Brazil’s Electoral Public Prosecutor subsequently recommended daily penalties, arguing that Meta failed to meet certain reporting requirements within the court’s deadlines. Studies conducted during the elections showed limits of existing safeguards. One examined more than 7,600 presidential campaign advertisements in Meta’s public Ad Library, and found indications that Meta charged the Lula and Flávio Bolsonaro campaigns different prices to reach voters during the first round. While there was a measurable disparity between the two campaigns, there was no way for outside observers to tell whether it reflected normal dynamics, technical failures, or unequal treatment.
Meta has denied deliberately favoring either candidate, attributing reported discrepancies to technical errors and gaps in its safeguards. Allegations of partisan manipulation have not yet been established in a final judicial ruling. After his victory, Lula pledged to regulate digital platforms, citing outside interference, in the election.
Meta’s role in Brazil is particularly significant because its influence extends beyond deciding which political content remains online. Unlike more traditional advertising channels such as television and radio, digital platforms distribute political advertisements through automated auctions, where costs and audience reach depend on advertiser bids, targeting criteria, and algorithmic predictions of engagement. As a result, two campaigns spending similar amounts can reach substantially different audiences, even when the same rules apply. These differences do not necessarily indicate political bias, manipulation, or interference, but they make it difficult to distinguish legitimate advertising dynamics from inconsistent enforcement or technical failures. The concern is especially significant during elections, when temporary advertising restrictions or changes in content distribution can affect campaigns within hours. Brazilian electoral authorities can require platforms to apply their rules equally, but enforcement by corporate entities is difficult.
The implications of Brazil’s experience extend beyond its electoral system. As political campaigns increasingly depend on digital platforms to reach voters, private technology companies have gained considerable influence over how political information is distributed. At the same time, electoral authorities often lack the tools to investigate their decisions. This creates a difficult imbalance: governments remain responsible for ensuring fair elections, but the companies control most of the information needed to assess whether platforms comply with electoral rules. The challenge also matters for other democracies trying to regulate digital political activity. The European Union, for example, has already implemented measures to address some of these challenges through its Digital Services Act, which requires major platforms to assess the risks to electoral processes and provide greater algorithmic transparency. In contrast, the United States lacks comprehensive federal regulation, intensifying concerns about AI-generated political advertising. States have been left to adopt their own approaches, requiring disclosure of AI-generated content or restricting manipulated media during specific time periods before an election.
Electoral authorities have strong reasons to require greater transparency from digital platforms, but doing so requires recognizing that equal rules mean little when companies cannot verify compliance themselves. Effective oversight requires not only clear rules but also independent mechanisms to assess whether platforms apply them consistently. This includes greater access to data on content distribution and recommendation systems, transparent reporting on account restrictions and algorithmic changes during election periods, and procedures for promptly investigating complaints from candidates and political parties.
Andre Nicola is a Program Coordinator for Latin America at ORF America and Caroline Arkalji is a Junior Fellow at ORF America.