Smile, You’re Being Watched by Brazil’s Judiciary

Courts, NGOs, and private companies have built a data-collection network that turns opinions and private messages into political targets.

Since the publication of the Twitter Files Brazil and Vaza Toga investigations, it has become increasingly evident that Brazil’s judiciary has assembled a parallel digital-surveillance system operating on an industrial scale. If Watergate brought down a president through the wiretapping of a single committee, what exists today in Brazil is far broader — an opaque and politically unaccountable apparatus run by courts, public agencies, and private firms.

Millions of Brazilians may already be included in secret databases, categorized by their opinions, criticisms, or even memes shared in private chat groups. The true scale of this system is impossible to measure: much of it operates in the shadows, through hidden algorithms and confidential platforms. How can we be sure the family WhatsApp chats aren’t being monitored right now?

The Censorship Industrial Complex is a direct descendant of the U.S. Military-Industrial Complex and the post-9/11 surveillance paranoia that fueled the so-called “war on terror.” Built to track external enemies, the security machine expanded without limits. After Donald Trump’s election in 2016, it was turned inward — against supposed “domestic threats” — under the banner of fighting fake news and “foreign interference.” What began as counter-terrorism evolved into a global system of citizen surveillance. Parts of it were later exposed by WikiLeaks and by Edward Snowden’s leaks — reported worldwide by Pulitzer-winning journalist Glenn Greenwald. But it was fully laid bare by the Twitter Files revelations published by Michael Shellenberger, Matt Taibbi, and other investigative journalists.

In Brazil, that model was imported by the Supreme Federal Court (STF) and the Superior Electoral Court (TSE). These institutions began to enlist tech companies, NGOs, and informal operatives to collect sensitive data, track social networks, and turn personal interactions into political intelligence reports. Public comments and private group messages on WhatsApp or Telegram were analyzed at scale and forwarded to judiciary-linked units. Any activity could be flagged as suspicious — from a sarcastic emoji to a tweet criticizing a Supreme Court justice.

As previously shown in A Investigação, one of the central tools of Brazil’s “juristocracy” is precisely this mass extraction of citizens’ data. It creates an invisible social map — later weaponized to justify prosecutions and censorship.

This report uncovers how, through a series of cases and leaks, a permanent monitoring system has taken shape — one that functions as a digital police state, legitimized under the pretext of fighting “disinformation” and “hate speech.”

Chapter I: The Private Mechanism

“If your brand were mentioned today in a group of 10,000 people, would you know? Not every conversation is public — and not every crisis starts on the feed. Some begin in the family group, others in leaked voice notes. The question isn’t if they’re talking about you, but: are you listening? At Palver, we monitor where no one else is looking.”

“Information is power — but only when you understand what’s behind it. The truth hidden in the data isn’t always clear, and that’s what we do: we decode chaos and deliver intelligence.”

These quotes, posted on Instagram by the digital-analysis startup Palver, sound like standard marketing copy. But behind the slogans, the company built — by its own account — a system capable of monitoring more than 100,000 WhatsApp groups and other platforms in real time, turning personal conversations into data for courts, fact-checking agencies, and government partners.

In late August, the Vaza Toga series continued through a new set of reports published by Revista Oeste, written by journalists Edilson Salgueiro, Rachel Diaz, and Carlo Cauti. The stories revealed the existence of a “parallel cabinet” allegedly run by Supreme Court justice Alexandre de Moraes to target political opponents. They also exposed how private companies and NGOs were enlisted to monitor, classify, and shape online political narratives.

According to the leaked material, surveillance was carried out by a consortium of NGOs, analysts, and private firms working directly with Moraes’s office. During Brazil’s 2022 elections, the TSE tracked keywords such as “auditable vote,” “I authorize,” “fraud in the voting machines,” and even the names of Supreme Court justices like Moraes and Luís Roberto Barroso. The order: trace, map, and report these expressions across social media and messaging apps — all in the name of combating disinformation.

Among the TSE’s partners were the NGO Instituto Democracia em Xeque (“Democracy in Check”) and the tech firm Palver Consultoria, which supplied the TSE’s Special Unit for Counter-Disinformation (AEED, its Portuguese acronym) with access to databases containing messages extracted from hundreds of thousands of WhatsApp, Telegram, Gettr, and other groups.

Over 100,000 Monitored Groups

During Brazil’s 2022 elections, Palver signed a cooperation agreement with the country’s Superior Electoral Court (TSE), granting the court direct access to its digital-monitoring dashboard. In practice, this meant that citizens’ conversations — especially from Bolsonaro-aligned right-wing groups — were tracked, tagged, and delivered to the court to justify political and legal actions.

The agreement provided Palver’s platform to the TSE free of charge, imposed confidentiality over all exchanged information, and required on-demand reports — but set no clear limits on the scope of data collection, nor any mechanism for public transparency.

Cooperation agreement between Palver and TSE

Palver analysts reported in real time the growth of content critical of the electoral system, Brazil’s electronic voting machines, and the Supreme Court itself. On September 24, 2022, for instance, an internal alert warned that the term “auditable vote” was reaching a “historic peak” on social media — raising alarms among court operators just days before the first round of voting.

At that time, Brazil was locked in an institutional clash between President Jair Bolsonaro’s administration and the judiciary over the push for paper-trail voting, a long-standing demand from Congress and large parts of society but repeatedly rejected by the STF and the TSE. The topic later became central to a Federal Police narrative portraying Bolsonaro as plotting a coup to contest the election results.

Leaked Vaza Toga messages show that surveillance was organized directly from Justice Alexandre de Moraes’s office. His chief of staff, Cristina Yukiko Kusahara, supervised operations. Thiago Rondon, an analyst tied to the NGO Instituto Democracia em Xeque (or “Democracy in Check,” DX), requested that expressions such as “fraud in the voting machines,” “state of defense,” and “annul the elections” be added to the monitoring queries.

Alongside Palver, DX tracked social-media narratives critical of the STF, the TSE, and the electoral process. Funded by entities such as the Heinrich Böll Foundation (linked to Germany’s Green Party) and groups from the Open Society network, DX counts Neca Setubal — heiress to the Itaú banking family — among its board members. Former U.S. State Department official Mike Benz has described DX as a bridge between U.S. intelligence agencies, USAID, and censorship-related programs in Brazil. The institute’s left-leaning stance is explicit in public comments by Beto Vasques, DX’s Institutional Relations Director, who has labeled right-wing lawmakers as “militias” and “mafias” and defended the removal of pro-Bolsonaro legislators from office.

This apparatus remained active through July 2023, spanning the January 8th riots in Brasília and connecting directly with the surveillance framework exposed by Vaza Toga and Twitter Files Brazil. What began as a marketing-oriented tool evolved into a political weapon — not to inform citizens, but to monitor and shape them.

The leaked materials show that Palver was used by the TSE to monitor WhatsApp and other platforms, flagging “harmful” content — including criticism of Supreme Court justices, calls for September 7th demonstrations (Brazil’s Independence Day), and discussions about paper-trail voting. The data were sent straight to the court.

Palver supplied the TSE with real-time dashboards, alerts, and analytical reports tracking posts, videos, voice messages, and mentions of sensitive terms. The queries — lists of keywords — were provided by court collaborators and included phrases like “electoral fraud,” “military intervention,” and the names “Barroso,” “Fachin,” and “Moraes.” In effect, any mention of these could be captured and flagged as potentially “dangerous.”

Technically, Palver could index messages by keyword, identify the groups where they circulated, and alert the court about spikes in mentions or narrative trends. Screenshots show that by September 1, the company had already warned the court about rising positive mentions of September 7. On Independence Day itself, Palver recorded a surge of messages supporting Bolsonaro and criticizing the STF. By September 28, it reported an explosion of posts discussing the voting machines.

The Rise of the Monitoring Machine

Earlier versions of Palver’s own website boasted of monitoring over 75,000 groups, integrating data from WhatsApp, Telegram, TikTok, X (Twitter), Reddit, radio, and television. It claimed operations in more than 10 countries, collecting data in 200 languages, tracking 2 million users, and processing 300,000 messages and 150,000 media files per day. Later, that figure grew to “over 100,000 WhatsApp and Telegram groups.”

Palver presents itself as a data-intelligence and digital-narrative company, helping governments and corporations “understand social behavior in the information age.” This kind of activity — known as social listening — is common in marketing and public-relations work, where agencies track public posts to gauge reactions to products or campaigns.

What sets Palver apart is its focus on private messaging apps like WhatsApp and Telegram. Unlike open social networks, these are considered closed, private environments, where data are not publicly available for scraping. Extracting messages from these groups without consent represents a serious threat to privacy and free expression, especially when done in partnership with state institutions.

The practice also violates Meta’s official policies. WhatsApp explicitly forbids automated data collection or surveillance without user consent. According to the company’s own terms of service: “You may not use our services in any way that involves automated data collection from our users without permission.”

Palver was founded in 2021 by Felipe Bailez, an economist with a background in finance, and Luis Gustavo Fakhouri, a public-administration specialist. Both write a weekly column called “Forwarded Frequently”, for Folha de S.Paulo newspaper, which turns their monitoring findings into snapshots of Brazil’s digital behavior.

In practice, however, that column has become a platform for narratives often favorable to the judiciary or the political left. Attorney and city councilman Rodrigo Marcial (Novo Party), creator of the Dossiê Moraes website, argues that Palver combines political bias and methodological opacity, operating as a black box within Brazil’s digital-monitoring ecosystem.

List of articles written by Palver's founders

Marcial notes that the company barely hides its ideological leanings: “The founders write weekly about ‘disinformation’ — from their own point of view. In other words, those who monitor also craft the narrative,” he wrote on X. The tone of their columns, he says, routinely favors the left and the judiciary, reinforcing perceptions of alignment.

One example cited by Marcial was an article in which Bailez and Fakhouri highlighted supposed praise from Brazilian neo-Nazis for right-wing congressman Nikolas Ferreira, after an attack on U.S. commentator Charlie Kirk. According to Marcial, the piece became ammunition for mainstream outlets to link Brazilian conservatives to extremism: “They label, they push bans, all under the ‘disinformation’ seal. Now they’re using the Charlie Kirk case to paint all of Brazil’s right as neo-Nazi. It doesn’t stick,” he wrote.

While Palver publicly claims to study communication trends and viral narratives, its tools effectively map the ideological color of digital conversations. In practice, its systems have been used to profile citizens’ political leanings — particularly among Bolsonaro supporters — fueling the creation of left-aligned narratives that serve Brazil’s ongoing culture and information wars.

The bias is evident in a statement made by Felipe Bailez during a debate held in April 2024. On that occasion, the CEO of Palver said that “the Brazilian right is extremely efficient at communicating — unfortunately, light-years ahead of the left when it comes to communication on WhatsApp.”

Contacted by our team about the revelations from Vaza Toga 3, Palver did not respond by the time this edition went to press; the space remains open for comment.

Exporting the Mechanism

In December 2023, Palver established a company in Delaware, USA — a state well-known for offering regulatory flexibility and tax advantages. This new entity became the official partner of Palver Brazil, enabling international cooperation within the global network of so-called “disinformation-combat” initiatives.

The following year, Palver expanded its global presence by coauthoring a study published in the Harvard Kennedy School Misinformation Review alongside American researchers. The paper, based on data supplied by Palver, analyzed the spread of content from Bolsonaro supporters across WhatsApp, Twitter, and the press during the events of January 8, 2023 — when pro-Bolsonaro demonstrators stormed government buildings in Brasília. The methodological appendix of the study offered rare insight into the company’s internal “engine”: mass monitoring, message harvesting, and subjective interpretation of narratives.

Palver’s technology soon appeared in collaborations with foreign organizations, such as the Digital Democracy Institute of the Americas (DDIA), based in Washington, D.C. The DDIA was founded by Roberta Braga, a former fellow at the Atlantic Council — identified by journalists Matt Taibbi and Michael Shellenberger as a key hub in the Censorship-Industrial Complex. Braga previously worked at Equis Research, a U.S. think tank that studies Latino public opinion and designs narrative campaigns around elections, immigration, and progressive policies.

Focused on tracking discourse within Latino communities in the U.S. and Latin America, the DDIA began using Palver’s datasets to map “sensitive narratives” circulating on WhatsApp and other platforms. In 2024, the institute reported that data supplied by Palver helped it identify over 3,200 false or misleading messages across 1,400 WhatsApp groups with Latino participants in the U.S. These findings were integrated into narrative-analysis reports, YouGov public-opinion studies, and “community intervention” programs — including training sessions for Latino journalists on topics such as elections, immigration, and environmental policy.

These partnerships place the Brazilian data-analysis firm at the center of transnational initiatives to monitor and shape public discourse, linking its operations directly to the infrastructure of the Censorship-Industrial Complex. The use of its technology by foreign organizations connected to political parties, billionaire foundations, and digital-influence agencies suggests that data captured inside Brazil — including from private citizen groups — may have been used abroad to guide perception-management strategies.

In Brazil, one of Palver’s main partners is Agência Lupa, a prominent fact-checking outlet that has worked with both the TSE and major media consortia during recent election cycles. The alliance between Palver and Lupa produced detailed reports on the January 8th riots, earning Palver the 2024 Innovation in Journalism Award for Combating Disinformation, granted by the International Center for Journalists (ICFJ).

The joint project tracked messages circulating in pro-Bolsonaro WhatsApp and Telegram groups, identifying — according to the authors — the reemergence of “coup narratives” and “calls for military intervention” five months after the January unrest. The reports also linked Bolsonaro-allied lawmakers to funding for buses that brought protesters to Brasília, reinforcing the narrative that the demonstrations were politically orchestrated. Lupa and Palver’s findings were promoted as valuable resources for journalists and government officials and showcased internationally by the ICFJ as an example of “investigative excellence.”

The ICFJ itself is funded by entities connected to the U.S. government, including USAID, the State Department, and the National Endowment for Democracy (NED) — an organization often accused of serving as a “soft power” arm for U.S. geopolitical interests in the Global South. Other major funders include the Open Society Foundations, the Bill & Melinda Gates Foundation, Craig Newmark Philanthropies, and tech giants such as Meta, Google, TikTok, YouTube, and Microsoft — all active players in global content-moderation programs. The ICFJ also acts as a liaison between the U.S. government and far-left organizations like Sleeping Giants Brazil.

In an official statement on the ICFJ website, Luis Fakhouri, Palver’s cofounder and Head of Strategy, celebrated the award:

“The ICFJ helped me by providing the network and the means necessary for me and my company, my network, and my colleagues to deliver the best possible work in combating disinformation.”

A “Digital Psychohistory”

In a video on his Instagram page, Fakhouri revealed that the company’s name, Palver, was inspired by a character from Isaac Asimov’s Foundation trilogy. In Asimov’s universe, Preem Palver is the leader of a secret society of intellectuals working behind the scenes to shape the fate of the galaxy through knowledge and psychological manipulation of the masses. Convinced that humanity left to its impulses would destroy itself, Palver considered mental and political manipulation a necessary evil — a way to preserve order and ensure civilization’s survival.

In Asimov’s story, this philosophy connects to “psychohistory,” a fictional science combining mathematics, statistics, and psychology to predict the collective behavior of large populations and steer history’s course.

The choice of this name carries an irony that’s hard to miss. Just like its literary namesake, Palver presents itself as a quiet intelligence, capable of understanding and anticipating collective behavior through data — not to comprehend, but to influence. In fiction, Palver symbolizes predictive control in the name of rational order. In reality, the company embodies a similar principle: a form of digital psychohistory applied to social networks and political trends — a science of prediction and persuasion, placed at the service of the State.

The Legal and Human Rights Dimension

Attorney Richard Campanari, a legal consultant specializing in Electoral and Civil Law and member of Brazil’s Academy of Electoral and Political Law (ABRADEP), considers the monitoring of closed WhatsApp and Telegram groups by companies contracted by courts to be extremely delicate from a constitutional standpoint.

He notes that the secrecy of private communications is explicitly protected by Brazil’s Constitution and can only be breached through a specific and well-founded judicial order. “If infiltration occurred in private WhatsApp or Telegram groups without such authorization,” he explains, “it would constitute a direct violation of a fundamental right.”

According to Campanari, even cooperation agreements with the Electoral Court (TSE) would not legally justify outsourcing such surveillance to private firms or NGOs. Any state initiative of this kind must rest on clear legal authorization, democratic oversight, and full transparency — requirements that, in his view, appear not to have been met.

“My criticism isn’t partisan — it’s constitutional,” he stresses. “When the State begins monitoring words or expressions like ‘auditable vote’ or simple criticism of public officials, it risks turning legitimate debate into suspicious speech.”

He warns that such practices promote self-censorship and undermine political pluralism, a foundational principle of Brazil’s republic. Any infiltration of private environments without a court order, he adds, is disproportionate and incompatible with democracy:

“By sweeping the internet for selected terms — none of which are defined as crimes — the State effectively creates a blacklist of forbidden words. That amounts to algorithmic censorship, inconsistent with free expression and public debate.”

Wallace de Oliveira, a legal consultant specializing in Digital, Civil, and Procedural Law, describes the Palver–TSE partnership as profoundly alarming. He argues that it represents the institutionalization of mass digital surveillance, with large-scale data collection and analysis of WhatsApp groups and social networks — resembling corporate business-intelligence systems but deployed for political purposes.

“It shows how tools originally designed for commercial analytics are being used for objectives that go far beyond legitimate administrative functions — amounting to possible censorship and ideological control of citizens,” he says.

Oliveira emphasizes that the issue goes beyond administrative law, touching both constitutional and international human-rights law. He cites possible violations of treaties ratified by Brazil, such as the American Convention on Human Rights (Article 13) — known as the Pact of San José, Costa Rica — which protects freedom of expression, and the International Covenant on Civil and Political Rights (Articles 17 and 19), which safeguard privacy and freedom of opinion.

From a digital-law perspective, he sees clear signs of private communication monitoring without consent, in breach of Brazil’s General Data Protection Law (LGPD) and the principle of informational self-determination. Administratively, he criticizes the lack of public transparency and competitive bidding in formalizing the partnership, and the potential abuse of purpose — “when the Judiciary assumes the role of filtering political narratives.”

On the human-rights level, Oliveira warns that the impact is “structural,” affecting the very core of human dignity:

“It undermines the freedom to think, to feel, to express, to disagree, and to question — all pillars of human dignity.”

The seriousness of the situation, he adds, is compounded by the fact that Palver is registered in the United States — a country that publicly positions itself as a defender of free speech. This connection could expose the company to U.S. jurisdiction under laws such as the Foreign Corrupt Practices Act (FCPA), the Global Magnitsky Act, or state-level privacy regulations like the California Consumer Privacy Act (CCPA), if evidence emerges that it aided human-rights violations abroad.

Oliveira also issues a historical warning, comparing the current climate of digital surveillance to past periods of social and political repression, when the mere fear of being watched or reported was enough to shape people’s behavior. “The difference,” he says, “is that today the State uses digital tools originally developed by private companies to exercise control, operating on the assumption that individuals lack the autonomy to govern their own actions.”

“The difference,” he says, “is that today the State uses corporate digital tools to exercise control, operating on the assumption that individuals lack the autonomy to govern their own behavior.”

He connects this mindset to recent Supreme Court precedents in Brazil, which have asserted that individuals can be deemed incapable of determining their own intent — effectively granting the Judiciary the power to define and restrict behavior.

“When such reasoning becomes generalized and automated, without clear legal limits,” he warns, “it directly violates fundamental rights: privacy, free expression, and due process.”

For Oliveira, the phenomenon mirrors Michel Foucault’s “panopticon” — a social structure in which constant surveillance is unnecessary because the mere possibility of being watched causes people to censor themselves.

“Neighbors, colleagues, relatives become potential informants,” he explains. “That changes not just public discourse but the very way society exists.”

He concludes that the convergence of digital surveillance, irregular delegation of state powers, and foreign collaboration represents a direct assault on the essence of the rule of law:

“As guardian of the Constitution, the Supreme Court must ensure that every administrative act meets the essential criteria of legitimacy — competence, purpose, form, reasoning, and object. When these are ignored, legitimate power turns into arbitrariness, leaving deep structural scars that will echo through future generations.”

Chapter II – The State That Watches

As revealed in The January Files (Vaza Toga 2), it was Alexandre de Moraes’s parallel office inside Brazil’s Electoral Court (TSE) that activated so-called “external partners” to scour WhatsApp and Telegram groups in search of evidence to justify the continued imprisonment of women detained after the January 8th riots in Brasília. Moraes’s chief of staff, Cristina Yukiko Kusahara, personally instructed Eduardo Tagliaferro, then head of the TSE’s Department for Combating Disinformation, to send the request directly to Moraes’s personal email, bypassing official channels — a clear indication that the operation was conducted outside the State’s formal structure. According to sources interviewed for this report, the minister personally authorized the use of these informal agents.

But they went even further. As exposed by A Investigação, Moraes’s informal team relied on infiltrators in private chat groups to support its operations. In a WhatsApp group called “Empresários & Política” (Businessmen & Politics), journalist Lucas Mesquita — who today holds a government-appointed position under President Lula — extracted conversations later used as the basis for an operation targeting business owners in August 2022, which included the collection of private information and the blocking of social-media accounts.

The material was leaked to the TSE by Letícia Sallorenzo, identified by our reporting as the so-called “Bruxa” (Witch). A close friend of Viviane Barci de Moraes (the minister’s wife, recently co-sanctioned under the Global Magnitsky Act), Sallorenzo routinely delivered dossiers directly to Moraes’s office and pressured platforms to remove targeted profiles.

Messages between Sallorenzo and Tagliaferro, obtained by A Investigação, reveal her eagerness to “get her hands” on a phone seized by the Federal Police — a device belonging to Meyer Nigri, CEO of the construction firm Tecnisa. Her goal was to access potential messages exchanged with Jair Bolsonaro via broadcast lists. Sallorenzo asked whether the phone was in custody of a “reliable Federal Police team” — a euphemism for agents allegedly aligned with Minister Moraes. Tagliaferro confirmed it was.

The Federal Police delegate Fábio Alvarez Shor was the key operational link in this chain. He signed the report that justified the raid — a document that, according to a forensic analysis commissioned exclusively by A Investigação, was backdated: although it bore the date August 19, it was actually created on August 29, several days after the operation had already taken place. The maneuver appears designed to create the illusion that the crackdown on businessmen was based on technical police findings, when in fact Moraes’s decision had relied solely on an article by journalist Guilherme Amado, published on the Metrópoles website.

The leaked conversations from Empresários & Política served not only as the pretext for the search-and-seizure operation ordered by Moraes but also as evidence in other judicial actions. The National Council of Justice (CNJ) opened a disciplinary inquiry against Judge Marlos Melek, accused of political participation in the group. Meanwhile, Father Júlio Lancellotti, a progressive activist priest, filed a defamation lawsuit against retail magnate Luciano Hang, citing statements attributed to him in the same chat group.

A separate but telling case illustrates how private messages are increasingly being judicialized. In 2023, then-Lulas’s Justice Minister Flávio Dino filed a lawsuit against a 66-year-old retiree over a WhatsApp comment made in his condominium group. The man had criticized Dino for “visiting and associating with organized crime and wanting to disarm the population,” referring to Dino’s visit to the Maré Complex — a Rio de Janeiro favela controlled by the Comando Vermelho, a Brazilian organized crime syndicate — and to his gun-control policies. The comment triggered a Federal Police investigation (under Dino’s own ministry) and eventually a criminal case, later closed through a settlement. In 2025, now serving as a Supreme Court justice, Dino filed a civil suit demanding R$30,000 (≈$5,500) in moral damages from the same retiree.

But it’s not only ordinary citizens who have faced retaliation. Whistleblowers who exposed the inner workings of this surveillance system have also been targeted. Among them is Eduardo Tagliaferro, the very official once responsible for coordinating anti-disinformation operations under Moraes.

After leaving Brazil, Tagliaferro sought asylum in Italy and began cooperating with journalists and authorities as a whistleblower, providing documents and testimony about pressure on tech companies and manipulation of internal TSE processes. In hearings before the Brazilian Congress, he presented materials showing altered case files and fabricated justifications for Federal Police raids against conservative figures — reinforcing allegations of an “ideological bunker” inside the Electoral Court and what he called a “persecution task force” against Brazil’s political right.

A Watched People

The Fake News Inquiries led by Supreme Court Justice Alexandre de Moraes, now entering their sixth year of procedural abuses and political lawfare, have evolved into a vast data-harvesting machinery. Hundreds of citizens have faced warrantless seizures, phone interceptions, and forced disclosure of digital data. Every conversation, banking transaction, and file extracted from seized devices has been turned into input for secret operations — executed by police delegates personally selected by Moraes.

One of the most controversial practices under his tenure is the fishing expedition — open-ended searches with no specific target, used to “fish” for evidence of crimes unrelated to the original warrant. Federal Police (PF) officials often justify them as “fortuitous discoveries” or “serendipity,” but in practice, they amount to reverse-engineered investigations designed to uncover new accusations after the fact.

The most emblematic case was the so-called “draft coup decree.” During a search of former Justice Minister Anderson Torres’s home, the PF claimed to have “accidentally” found a draft presidential decree proposing federal intervention in the electoral process. The police record itself admits there was no specific warrant authorizing the seizure of that document — making it an illegally obtained piece of evidence, the product of a fishing expedition.

The Twitter Files Brazil further revealed how this model of mass data collection extended to social media. The most notorious example was an STF order compelling X (formerly Twitter) to hand over the IP addresses and registration data of all users who created, liked, or retweeted posts containing hashtags such as #VotoImpresso (“paper-trail vote”) — roughly 41,000 posts. While the orders technically covered all sides, they overwhelmingly targeted conservative profiles.

Another case, reported by journalist Rodrigo Rangel on Metrópoles in January 2023, exposed a sweeping warrant authorized by Moraes. The order allowed investigators not only to surveil eight named individuals but everyone connected to them — potentially including politicians, journalists, judges, military officers, and even the president himself. The decision granted access to phone records, telematic data, and geolocation history dating back to 2017, encompassing Brazil’s entire pre-election period.

Later, in August 2023, under Moraes’s orders, the Federal Police began cross-referencing data to identify every individual who donated via Pix (instant payment app) to former President Jair Bolsonaro, totaling R$17 million (≈$3.3 million). The operation involved the banking and tax data of Bolsonaro and his wife Michelle, analyzed through the Simba financial-tracking system — tracing every CPF (tax ID) linked to donations. The official rationale was a suspected money-laundering scheme involving jewelry sales abroad. In practice, however, the move turned thousands of small donors into potential criminal suspects, despite their donations being legal. Bolsonaro’s defense team called it political persecution, noting that he had received nearly half of Brazil’s total votes — meaning the inquiry theoretically targeted millions of citizens.

In another related investigation — the Digital Militias inquiry — the Federal Police demanded that Meta and X provide the complete list of followers, along with all email addresses linked to his accounts, of conservative journalist Allan dos Santos, now living in exile in the US.

This surveillance model was institutionalized within the Electoral Court starting in 2022, with the creation of the Special Advisory for Countering Disinformation (AEED). Initially presented as a technical initiative to curb fake news, the AEED quickly evolved into a permanent data-monitoring mechanism. Under Moraes, the system began operating 24/7, with direct access to sensitive data and constant interaction with both tech platforms and state agencies. Influencers, journalists, and even ordinary citizens who criticized the court were monitored, with detailed reports sent straight to Moraes’s office.

That same year, Moraes also created a proprietary intelligence unit within the TSE, operating outside Brazil’s official intelligence framework. This unit had autonomy to produce reports, cross-reference databases, and contact platforms directly. Its stated purpose was “to guarantee the integrity of elections,” but in practice, it established a clandestine surveillance arm answerable only to the TSE president — with no external oversight or legal framework.

In March 2024, the TSE went further, launching the Integrated Center for Countering Disinformation and Defending Electoral Democracy (Ciedde). Publicly framed as an anti-disinformation initiative, Ciedde effectively expanded the digital-monitoring infrastructure pioneered by the AEED. The new center incorporated the Federal Police, Federal Prosecutor’s Office, Brazil’s Intelligence Agency (ABIN), and major tech platforms, operating as a permanent monitoring hub with real-time access to social-media data streams.

Ciedde’s official logo — an Orwellian “all-seeing eye” — symbolized what it had become: a centralized information-control hub embedded inside the Electoral Court itself. Staffed by contractors linked to NGOs and foundations aligned with the government, the center functioned as a continuous surveillance room, receiving anonymous complaints, filtering content, and ordering takedowns without public transparency — all under the banner of “defending democracy.”

The “January 8 Certificates”

The post-riot crackdown on January 8 detainees did not follow traditional legal procedures. Instead of investigations led by the police or prosecutors, the search for information was directed from within the Supreme Court, through a parallel intelligence cabinet established under Moraes’s direct orders.

The process began with lists from police agencies, containing names, ID numbers, photos, and raw data of detainees. These lists were forwarded informally to ad hoc teams within the judiciary. From there, officials connected to the STF and TSE accessed confidential internal databases — including GestBio, the Electoral Court’s biometric system, which stores fingerprints, photographs, signatures, and other sensitive data on nearly all Brazilian voters.

The objective was not to verify criminal records or identify violent offenders, but to put faces to names. Once a photo was retrieved from GestBio, it was cross-referenced with data from the Federal Revenue Service and Renach (Brazil’s driver database) to locate individuals’ social-media accounts. Investigators then scoured those profiles for any post or comment deemed “anti-democratic.”

There were no clear criteria. A “like,” a meme, or a sarcastic remark about Lula or the Supreme Court could be enough to generate what insiders called a “positive certificate” — an informal label used to justify extended detention, even without access to the alleged evidence by the defense.

One of the most absurd cases was that of Ademir Domingos Pinto da Silva, a 54-year-old street vendor who did not even participate in the government-building invasions. He arrived in Brasília hours later to sell flags but was arrested anyway and flagged as “positive” based on old tweets from 2018 criticizing the Workers’ Party (PT) and Lula — four years before the election and five years before the supposed coup attempt.

That “certificate,” signed by the TSE’s Disinformation Unit, was enough to keep him imprisoned for four months, later forcing him to wear an electronic ankle monitor and attend a mandatory “democracy reeducation course.”

The Lula Government’s Role in Surveillance

The Lula administration has expanded and institutionalized the surveillance framework originally developed by Brazil’s judiciary. One of its flagship initiatives was the creation of the National Attorney’s Office for the Defense of Democracy (PNDD) — widely nicknamed the “Ministry of Truth” — within the Attorney General’s Office (AGU), led by Jorge Messias. This new branch effectively turned the State’s legal apparatus into a speech-control mechanism, initiating investigations against citizens, journalists, and influencers accused of spreading “disinformation.” The PNDD began issuing extrajudicial notices, demanding public retractions, and filing lawsuits to remove content critical of the government — actions far beyond the AGU’s traditional constitutional mandate of defending the Union in court.

Meanwhile, Serpro, Brazil’s federal data-processing agency, launched the Digital Citizen Project, whose public call describes its goal as delivering “relevant information to the right people, at the right time,” with full traceability and feedback loops. Officially framed as a transparency and engagement tool, the program in practice enables the collection of sensitive personal data and the delivery of personalized political messages — a sophisticated form of state propaganda disguised as civic service.

The recent record of Lula’s Secretariat for Social Communication (Secom) reinforces this trend of politicized technological management. In March 2024, the ministry opened a R$197.7 million (≈$38 million) bidding process to hire marketing firms for digital communication strategies, social-media management, and engagement tracking. In July 2024, Brazil’s Federal Audit Court (TCU) suspended the process after uncovering irregularities — including leaked technical proposals, violations of confidentiality, and favoritism. The contract was later reissued in January 2025 for a reduced amount, then relaunched in July for R$98.3 million (≈$19 million), involving three agencies with similar mandates.

During the Rio Grande do Sul floods, Secom coordinated groups known as “Fake News Hunters”, organized to flag and suppress posts criticizing the government’s response to the disaster. Officially labeled as volunteers, these groups operated under the direct coordination of staff linked to Minister Paulo Pimenta, amplifying government talking points while targeting opposition accounts. The case exposed the merging of state-funded communication, political propaganda, and online surveillance, consolidating a model in which public institutions use taxpayer-funded digital activism to shape discourse and silence dissent.

Digital Identity, Surveillance, and Control

While Europe and the United Kingdom are engaged in heated debates over the risks of digital identity systems and automated population monitoring, Brazil has already quietly implemented one of the world’s most advanced — and least supervised — data-driven systems of state control.

In Europe, the eIDAS 2.0 regulation (the European Digital Identity) came into effect in 2024, introducing the EUDI Wallet — a standardized digital ID for all EU member states. It is now mandatory for accessing public services and stores citizens’ biometric data. The measure has provoked strong opposition from civil-liberties groups such as European Digital Rights (EDRi), which warns of the growing risk of mass surveillance and erosion of individual freedoms.

In the United Kingdom, Prime Minister Keir Starmer’s Labour government announced in September 2025 the creation of a mandatory digital ID for workers, informally dubbed the BritCard. The card includes health data, AI-based authentication, and integration with employment systems. The proposal alarmed even members of Starmer’s own party, who compared it to China’s social credit system, warning of “total digital control of the workforce.”

Journalist Michael Shellenberger harshly criticized the British plan in an article for Public News, arguing that its true purpose extends beyond efficiency or modernization: it aims at total surveillance, censorship, and behavioral control. He cited the involvement of billionaires like Larry Ellison (Oracle), Bill Gates, and Tony Blair — whose institute received £257 million to centralize public data through AI. Shellenberger wrote that Ellison himself once boasted, “Citizens will behave better because we’re recording everything.” He denounced the manipulation of research by the Tony Blair Institute, warned of the dangers of data leaks and hacking, and drew direct parallels with China’s digital control model. According to Shellenberger, this agenda is driven by corporate interests worth billions, while the silence of NGOs exposes the hypocrisy behind the so-called “digital rights” movement.

In an interview with Quillette, Shellenberger suggested that disinformation laws, digital ID programs, and political censorship may be interlinked as part of a broader trend toward “digital authoritarianism.” He warned that new disinformation laws could criminalize legitimate political speech, while digital identity systems would serve as tools for tracking, scoring, and disciplining citizens.

Although these initiatives in Europe and the U.K. are still under debate or early implementation, they are presented as symbols of technological progress. But what many Europeans still fear as a dystopian future — a centralized system of identity, authentication, and behavioral tracking — is already operational in Brazil.

In 2023 the Lula government officially designated the CPF (Individual Taxpayer Registry) as the single national identifier and introduced the National Identity Card (CIN) — a unified document containing a QR code and chip with biometric data, mandatory by 2032. By October 2025, more than 30 million CINs had already been issued. The centralized database is managed by gov.br, Brazil’s digital government platform, which integrates over 4,500 public services, many requiring facial authentication and cross-referencing of health, income, and employment records.

But in Brazil, digital identity is not just a convenience — it has become a precondition for civil existence. Access to social programs, labor benefits, student loans, medical exams, and even civil-service exams requires registration and authentication through gov.br. A citizen who refuses to use the platform risks becoming invisible to the State.

This model of control is already in full operation. As revealed in Vaza Toga 2, Brazil’s judiciary holds privileged access to sensitive databases such as the GestBio system, managed by the Electoral Court (TSE), which stores the biometric data of millions of Brazilians — collected compulsorily under the pretext of electoral security. Officially designed to ensure that each voter votes only once, the system was granted to the Federal Police by order of Justice Alexandre de Moraes to help identify suspects involved in the January 8th riots. One man, who had never been to Brasília, was imprisoned based on old political posts and a “biometric similarity match.” What was once meant to protect the vote is now used to justify imprisonment.

At the same time, Brazil is testing its central bank digital currency (CBDC), known as Drex, which promises to modernize the financial system but, in practice, extends state control over citizens’ economic lives. Every transaction will be traceable and tied to a verified identity, allowing authorities or financial institutions to program or restrict payments according to predefined rules. The Drex pilot program has already sparked public backlash and a growing #DrexNão (“No to Drex”) campaign on social media. Congresswoman Júlia Zanatta called it a “tool of financial enslavement.”

Meanwhile, Brazil’s Censorship-Industrial Complex has capitalized on the scandal of the so-called “Parallel ABIN” — a small-scale intelligence unit from the previous administration — to accuse Bolsonaro’s government of political espionage. Its main focus was a basic tracking program known as First Mile, which triangulated cellphone tower data to estimate user locations. While the tool was misused — a fact investigated by then-director Alexandre Ramagem — it neither intercepted messages nor accessed private communications. Its scope was minimal, especially when compared to the mass surveillance infrastructure now operated by the Judiciary, which is capable of transforming ordinary online interactions into political evidence.

The contrast is stark. The same organizations and public figures who once denounced the supposed “techno-authoritarianism” of the Bolsonaro era now remain silent in the face of far more intrusive measures imposed by the Supreme Court, Electoral Court, and the Lula administration.

During the previous government, groups like Data Privacy Brasil, LAUT, and Conectas produced international reports, filed human-rights complaints, and campaigned against state surveillance. Today, despite the extensive revelations of Twitter Files Brazil and Vaza Toga, these same voices are mute.

This selective silence reveals more than mere ideological bias — it marks a shift of allegiance: from watchdogs of digital freedom to silent collaborators in the new architecture of control. The once-defiant language of “privacy rights” has been replaced by the bureaucratic jargon of “data governance” and “digital inclusion,” while mass censorship and surveillance are rebranded as “the fight against disinformation.”

Thus, Brazil — once praised as a model of civic resistance to technological authoritarianism — has become a laboratory for the normalization of digital surveillance, now carried out with the blessing of those who once claimed to oppose it.

Chapter III – The Police State and Brazil’s New Stasi

In the German Democratic Republic (GDR), the feeling of being constantly watched was not paranoia. The Stasi, East Germany’s feared secret police, built one of the most extensive surveillance networks in modern history. By the late 1980s, it employed around 90,000 full-time officers and nearly 200,000 informants. In a population of just over 16 million, that meant one collaborator for every sixty citizens. Co-workers, neighbors, even childhood friends could secretly report on one another.

Every conversation could be recorded. The Stasi maintained a dedicated division for telephone interception, obsessively monitoring the small segment of the population that owned a phone. Entire workshops were devoted to crafting hidden microcameras, button-sized microphones, and other devices capable of turning an ordinary living room into a covert listening post.

Only after the fall of the Berlin Wall in 1989 did the true scale of this machinery of control come to light: more than 100 kilometers of paper archives, along with millions of photographs, tapes, and film reels. Investigators also found millions of shredded document fragments, the remains of desperate attempts to destroy evidence as the regime collapsed.

That world is gone — but the logic that sustained it has not disappeared. The same impulse to watch, classify, and control has merely evolved with technology. And this logic, now digitized, is being dangerously replicated in modern Brazil.

Technology companies, NGOs, informants, and even judicial bodies have joined forces to collect sensitive data on millions of citizens, monitor private WhatsApp groups, and convert personal interactions into political intelligence reports.

Digital surveillance is no longer the exception — it has become the method. What once seemed like a dystopian relic of the Cold War now operates as official state policy, justified by a new slogan for old authoritarian instincts: “the fight against disinformation.”