Beyond hacking: How foreign governments can exploit vulnerabilities to manipulate elections

Research shows that electoral threats do not only affect the ballot box, but also hijack public debate, using polarization to erode trust in democratic institutions.

Ads

A recent report by the Brazilian Intelligence Agency (ABIN) pointed to the possibility of direct interference by the United States in the upcoming Brazilian elections, the first round of which will take place on Oct 4, 2026. According to the document the U.S. government, in an effort to curb China’s influence in Latin America, is stepping up efforts ranging from diplomatic pressure to possible campaign financing and the promotion of social media profiles.

This assessment gained traction in late September when U.S. Democratic senators questioned Secretary of State Marco Rubio about President Donald Trump’s stance on the Brazilian election. The lawmakers demanded that the U.S. government would recognize the results.

As researchers who closely monitor political communication, we see in these episodes a concrete example of a phenomenon we have investigated in depth in recent years. Of course, one country’s attempt to influence another’s domestic politics is nothing new. Researchers Paul Baines and Nigel Jones argued in their 2018 article Influence and Interference in Foreign Elections, that such interference has always existed. The difference is that digital platforms have given it unprecedented speed and concealment.

But how does such attempts at interference play out in real time? To answer this question, we analyzed hundreds of articles and intelligence reports, cross-referencing what academics and defense agencies already know about the use of digital media as diplomatic weapons. This work is part of a master’s thesis recently defended at the Graduate Program in Social Communication at PUC-Rio.

When trying to understand how these mechanisms work, the first mistake is to imagine a movie-like scenario. When people talk about foreign interference in electoral processes, most think of a technical attack, such as a hacker stealing passwords or tampering with the numbers on an electronic voting machine.

Our research shows, however, that the target of these attacks is not machines, but rather the public debate. The article Hacking Democracy, Managing Influence Campaigns and Disinformation in the Digital Age, by Niels Schia and Lars Gjesvik, illustrates this point well. The authors conclude that the goal of modern cyber influence campaigns is not to bring down computer systems, but rather to erode trust in democratic institutions and shape citizens’ perceptions.

By cross-referencing studies and security documents, a pattern becomes clear. Foreign interference does not occur only in the run-up to elections, but also on an ongoing and “hybrid” basis. This means that external actors combine digital tactics, such as automated accounts and the cloning of news portals.

For example, the European Union’s diplomatic service’s Third Report on Foreign Information Manipulation and Interference Threats, published in March 2025, recorded 505 incidents of external digital threats in 2024 alone. These incidents involved 68,000 pieces of content published across 38,000 channels on various platforms, with a focus on rapidly spreading disinformation.

In addition, many ads are designed to manipulate emotions and prompt real world actions. As W. Lance Bennett and Steven Livingston detail in the article Platforms, Politics, and the Crisis of Democracy, platforms are designed to capture attention by maximizing emotional engagement. They become an invisible battlefield between countries.

Sowing discord

It’s comforting to blame only foreign intelligence or bots when political debate descends into chaos. But the main finding of our investigation led us to conclude that manipulation from outside does not create confusion out of thin air. It tracks down and pours fuel on divisions that already exist within the country.

The RAND Corporation 2021 report Countering Foreign Interference in U.S. Elections helps explain this dynamic. In assessing U.S. elections, the document revealed that memes promoted by Russian agents circulated effectively by exploiting identity-based divisions in American society, such as the debate over gun ownership.

Campaigns by foreign actors rely on political, identity-based, and social fractures within each society. If a country already suffers from polarization, radicalized voters, and growing distrust in political institutions, it provides a fertile environment for external narratives to be embraced as if they were legitimate uprisings.

The ultimate goal is not merely to ensure a candidate’s victory at the polls. The greatest victory for those orchestrating these attacks is to plant the seed of doubt. By exploiting these fissures, these campaigns generate a slow and continuous erosion of trust in the democratic process.

Different approaches, a common challenge

During our research, we noticed a gulf between how universities study this problem and how government security agencies deal with it on a day-to-day basis.

On one hand, researchers are concerned with long-term harm. A 2024 study led by Ceren Budak, titled Misunderstanding the Harms of Online Misinformation_, argues that disinformation saturates the environment with noise and destroys what scholars call the “epistemic infrastructure”. The pollution of public debate by social media is so pervasive that society loses its basic ability to agree on which facts are true and, consequently, to make informed democratic decisions.

On the other hand, reports from government agencies — which dictate defense policies around the world — adopt a stance of tactical urgency. Governments prioritize the pragmatic approach of identifying inauthentic networks, cutting off funding, and forcibly containing immediate threats, such as the flood of AI-generated fake images and profiles.

The big problem is that these two perspectives almost never engage in an integrated dialogue. Furthermore, our analysis revealed that both scientists and intelligence agencies stumble over the same obstacle when trying to determine the true extent of the damage. Today, it is still virtually impossible to measure whether these multimillion-dollar disinformation campaigns actually sway votes at the polls — and to what extent.

A study looking at how Russian used social media to influence the 2016 U.S. election, conducted by researcher Gregory Eady, highlights this difficulty. In analyzing the Russian influence campaign on X during the U.S. elections, his team found that, although millions of users were exposed to manipulated content, it was not possible to prove that this led to a shift in votes.

Why does this happen? The answer exposes the greatest power asymmetry of the digital age. Research relies, to a large extent, on data provided by the big tech companies themselves. Researchers and governments operate in the dark, navigating through opaque algorithms and private platforms that guard their information as trade secrets. Without transparent access to the data architecture, we are unable to accurately measure how voters change their minds after being bombarded with micro-targeted ads.

We also encountered massive geographical bias while attempting to understand the problem. Throughout our research, we found that the overwhelming majority of studies, investments, and security tools are focused on what is happening in the United States and Europe. Countries in the Global South suffer the severe consequences of these external operations, yet their realities continue to be interpreted through the lens of the Global North’s concerns.

Shielding Democracy

Defending elections in the 21st century requires understanding that technology, by itself, does not create democratic fragility. To address this scenario that threatens national sovereignty, the debate must move beyond the exclusive logic of cybersecurity and the isolated fight against fake news.

Effectively protecting electoral integrity requires combining the critical insights of academia with the pragmatism of government agencies. Confronting these threats involves reducing our own vulnerabilities and building a more transparent digital ecosystem.
Gustavo Magalhaes da Silva Junior received funding from CNPQ while pursuing his master’s degree at PPGCOM, PUC-Rio.Arthur Ituassu is an Associate Researcher at the National Institute of Science and Technology in Digital Democracy (INCT.DD) and a Board Member of the Democracy in Check Institute (DX). He receives funding from CAPES, CNPq, and FAPERJ. He is also a Visiting Professor at the Center for Latin American Studies at the University of Arizona (CLAS/UA) with funding from Fulbright.

Tags: Academic Journalism, The Conversation, The Conversation Academic News