Imagine standing at a crossroads where technology, politics, and democracy all collide. That is Brazil in the days before its 2026 general election on October 4. This is not just another election; it is a global test case for how artificial intelligence can be used to influence voters, bypass traditional media, and spread disinformation at a scale we have never seen before. Far-right populists around the world have learned to use AI as a weapon against the institutions that once filtered information, and Brazil has become their most important battlefield. The country has been shaken by political disinformation for years, and now generative AI has made the threat even more dangerous. Thirteen candidates are running for president, but the race is dominated by two figures: incumbent Luiz Inácio Lula da Silva, from the Workers’ Party, and Senator Flávio Bolsonaro, son of former president Jair Bolsonaro. Lula narrowly won in 2022, but the shadow of that election still hangs over the nation, because Jair Bolsonaro is now imprisoned for his role in plotting a coup to overturn the result. The stakes could not be higher. Brazil is not only a democracy under stress; it is also a strategic prize. The country is becoming a priority market for data-center investment and supplies critical minerals needed to build and power AI systems. Both the United States and the European Union are looking to Brazil as they try to reduce their dependence on China, and Flávio Bolsonaro has signaled support for that realignment. That means whoever controls Brazil’s government will help shape not just domestic policy, but the future of the global AI economy. Foreign governments and tech companies alike have a massive interest in the outcome, making this election a prime target for interference. What happens in Brazil over the next few days will send ripples far beyond its borders.
Brazil has long been a testing ground for information warfare tactics in the digital age. The country’s digital environment has been described as fertile ground for innovative strategies to spread polarizing content and manipulate public debate. Several factors make Brazil especially vulnerable: people rely heavily on hyperconnected chat applications, political polarization runs deep, and the media market is driven by powerful influencers who can reach millions in seconds. Brazil was also one of the first countries to recognize the dangers of AI in elections. Ahead of the 2026 election, the Superior Electoral Court, known as the TSE, issued rules prohibiting generative AI systems from recommending candidates, parties, or campaigns, and from expressing any electoral preference. These systems were also banned from producing content that could spread misinformation or bias. In theory, those rules should protect voters from AI-driven manipulation. In practice, they are not working. Research has found that most major AI models ranked or ordered candidates in nearly every instance during testing, and they often did not use official sources when doing so. Responses were inconsistent and changed based on how questions were worded, meaning different voters could receive completely different information with different levels of reliability. No one outside the companies themselves can see all these responses, so no one knows how much misleading content voters are actually being exposed to. Meanwhile, Meta’s advertising system has been approving ads that violate both its own community standards and Brazilian election laws. Watchdog groups like Eko found approved ads calling for a repeat of the January 8 attempted coup and others inciting violence against lawmakers. Some ads targeted lawmakers with hate speech, falsely celebrated Trump’s endorsement of Flávio Bolsonaro, and pushed the conspiracy theory that Jair Bolsonaro is a political prisoner. The rules meant to keep AI and platforms out of election manipulation are being broken openly, and voters have no way of knowing the full scale of the problem.
This has thrown Brazil into a major battle between state regulators and global tech giants. Brazil’s courts have imposed some of the most aggressive anti-disinformation measures in the world, including strict bans on deepfakes and rules that hold platforms accountable for the content they carry. But this top-down approach has turned the digital environment into a proxy war over sovereignty and free speech. High-profile clashes between Brazilian authorities and tech companies have become political weapons, used by far-right actors to fuel anti-establishment narratives and deepen public distrust. In 2024, after intense corporate lobbying, Brazil’s Senate removed proposed safeguards against high-risk AI models. A year earlier, Google and Telegram were investigated for allegedly interfering with Brazil’s Fake News Bill, which sought to make platforms responsible for misinformation. Big Tech has effectively forged an alliance with the far-right in Brazil, just as it has embraced the Trump administration in the United States, in order to avoid regulation. Elon Musk has been at the center of this conflict. In 2024, he challenged Brazil’s Supreme Court over its attempts to block X accounts associated with disinformation networks, and he actively encouraged Brazilians to use VPNs to bypass the blocks. The court responded by including Musk in its disinformation investigation and accusing him of obstruction. X was blocked entirely in Brazil in August 2024, though the block was lifted two months later after the platform complied with court orders. Musk’s confrontation made him a hero to the Bolsonaro movement. At a rally, Jair Bolsonaro praised Musk as “the man who really preserves true freedom for all of us,” tapping into a culture war that casts tech billionaires as defenders of liberty against corrupt courts. Then, in January 2025, Meta CEO Mark Zuckerberg joined the attack, insinuating that “secret courts” in Latin America were running a mass censorship campaign, widely understood as a reference to Brazil’s Supreme Court. Zuckerberg also promised to work with President Trump to push back on governments that go after American companies. The message was clear: the most powerful tech companies in the world are willing to fight Brazil’s regulators, and they have chosen sides.
The fight over AI and disinformation in Brazil is deeply connected to how voters get their information. Technology companies have influenced elections for years, both directly and indirectly. People search for news, share it on social media, and are then targeted by political ads built from data about their online behavior. In Brazil, the pattern has been especially troubling during the last two election cycles, with false claims spreading wildly across WhatsApp and other messaging platforms. Now, the concern has grown to include AI chatbots as primary sources of bias and false information. Like WhatsApp, AI chatbots deliver content privately. The public cannot see what they are telling voters, and auditors, election observers, and fact-checkers have no way to inspect it. During previous campaigns, political operatives in Brazil ran vast numbers of WhatsApp groups that appeared distinct but were actually controlled by a few administrators, hiding the true source of the political messaging. AI chatbots operate from a similar black box. Most users have no idea where a chatbot gets its information or why it chooses to emphasize one candidate over another. This marks a fundamental shift in how Brazilian voters are influenced. Increasingly, people are turning to private, highly personalized channels for political information, and those channels are invisible to anyone trying to ensure fairness. No one knows exactly how much mis- and disinformation voters are encountering in the run-up to the election, because no one can see what people are seeing. This is especially dangerous because digital platforms have repeatedly been shown to amplify right-wing propaganda disproportionately and to allow partisan bias to affect decisions about what content users can access. On September 23, just eleven days before the election, Meta temporarily prevented Lula’s campaign page from running political ads and removed an Instagram account that had reported on Flávio Bolsonaro’s alleged connections to financial scandals. Meta later reversed course, but the explanation was unclear, and the damage had already been done. Three years earlier, Google had used its Brazilian search homepage to direct users toward opposition to proposed platform regulation, encouraging them to lobby lawmakers. These are not neutral tools. They are active players in the democratic process, and their power is largely unchecked.
While the decision-making inside tech companies remains opaque, one thing is clear: AI-generated election disinformation has been spreading rapidly in the days and weeks before the election. In just the first eleven days of the campaign, the TSE received nineteen cases involving the use of AI in political advertising. Flávio Bolsonaro has been especially aggressive in using this technology. In July 2026, he released an AI-generated campaign video featuring deepfake footage of his father, who is under house arrest, making it appear as though the former president was out on the campaign trail with his son. Jair Bolsonaro is barred from making public statements or using social media, but the deepfake created the impression of a personal endorsement. The former president remains a popular figure in Brazil, and his endorsement is widely seen as important for his son’s campaign. Left-wing parties challenged the video, arguing that it illegally circumvented the restrictions imposed on Jair Bolsonaro. The Superior Electoral Court ruled, by a narrow four-to-three vote, that the video did not count as public electoral advertising because it appeared before Flávio’s candidacy was formally registered. The court then issued a broader definition of prohibited deepfakes for the rest of the campaign. But that did not stop Flávio. He has produced at least one AI-generated campaign song and shared several more AI campaign videos. Beyond the official campaign, supporters of Bolsonaro have reportedly organized massive digital “armies,” including a WhatsApp group where members receive daily “visual ammunition” to post on social media. A companion website contains ready-made posts, allowing users to filter content by social media platform or category. One of the largest categories focuses on criticism of the Workers’ Party. This mirrors previous astroturfed digital campaigns led by members of the Bolsonaro movement, the very campaigns that incited an attempted coup several years ago. The machinery of disinformation is not just alive; it has been upgraded with AI, making it faster, cheaper, and harder to track.
Whether these digital armies will try to fuel another insurrection remains unknown. But what is certain is that tech billionaires have an enormous stake in the outcome of this election, and they have powerful tools to shape what voters see. AI can put a sidelined leader back on the campaign trail, as Flávio Bolsonaro demonstrated with his deepfake video. Platforms can amplify a political attack or interrupt a campaign’s advertising, as Meta did to Lula in the final days of the race. The companies that control these tools are not neutral bystanders; they are fighting for access to Brazil’s government, its markets, and its resources. Whoever wins the upcoming election will help write the rules that tech companies must follow, both in Brazil and potentially around the world. The danger is that the people trying to win power and the companies helping them reach voters seem to find accountability equally inconvenient. For ordinary Brazilians, the consequences are real. They are being asked to make one of the most important decisions in their country’s history, but the information they rely on is increasingly generated by machines, filtered through secret algorithms, and delivered in private spaces where no one can verify its truth. The election in Brazil is not just about one country or one leader. It is about whether democracy can survive in an age where technology makes it possible to personalize propaganda for every single voter, hide the source of that propaganda, and do it all at a scale that human fact-checkers cannot hope to match. The world is watching, but the voters of Brazil are the ones who have to live with the consequences. They are navigating a new kind of political reality, and the rest of us may soon find ourselves in the same position.