es 
By ProBox and Linterna Verde
On June 21, 2026, Colombia held the closest presidential election in its recent history, which resulted in the victory of far-right candidate Abelardo de la Espriella amid high political polarization and questions about the integrity of the process.
Before and after the election process, allegations of fraud, doubts about the pre-counting software, challenges to the results, and accusations of foreign interference featured prominently in the public debate, especially in the wake of Gustavo Petro’s social media posts. These narratives transcended Colombia’s borders as vote counting and political reactions unfolded; state-run media in Russia and China, along with platforms in Venezuela and Argentina, prominently featured these events in their coverage for international audiences.
This investigation arises in that context—not to determine whether electoral fraud occurred, nor to evaluate Colombian media coverage, but to analyze how a specific set of international platforms presented the Colombian elections and which aspects of the process received the most visibility in their content.
To that end, ProBox, with the support of Linterna Verde, monitored coverage from eight media outlets over a 44-day period, from May 10 to June 23, 2026: teleSUR, Resumen Latinoamericano, Sputnik Mundo, Actualidad RT, TASS, CGTN, Xinhua, and China Daily. The sample included 372 posts published on websites and official accounts on X, Facebook, Instagram, YouTube, and TikTok. Collectively, the analyzed accounts have more than 252 million followers, a figure that demonstrates the broad potential reach of the content published by these media outlets on social media.
X represented 50% of the total sample (187 posts) and served as the primary platform for immediate reactions to key milestones in the process. The websites, with 69 posts, featured the most in-depth and technical content on election software and the debate between preliminary counts and final tallying, especially during the runoff election. On Facebook and Instagram, visual formats were prioritized, mainly on teleSUR and Resumen Latinoamericano; meanwhile, YouTube saw the most activity at the start of the campaign, and TikTok recorded only a single post throughout the entire period.
Latin American Block: Polarization and Electoral Suspicion
The Latin American block, consisting of teleSUR and Resumen Latinoamericano, was the most productive with 187 posts, accounting for 50.3% of the sample.
Of these, teleSUR was the most active outlet, with 153 posts. Furthermore, it accounted for 42.8% of all explicit mentions of the word “fraud” identified during the monitoring. Its coverage repeatedly highlighted President Gustavo Petro’s statements regarding the Thomas Greg & Sons software, Iván Cepeda’s complaints, and allegations of foreign interference as constant and corroborative evidence to fuel a narrative of technical and political fraud, despite these claims having been refuted by international election observation organizations.
This outlet also stood out for its speed and widespread presence: each news story was adapted into three or four different versions to maximize its impact on different audiences. While on X it functioned as a channel for immediate reporting, on Facebook and Instagram it prioritized visual content and more in-depth coverage; and on YouTube, the content was produced in English.
Meanwhile, Resumen Latinoamericano (34 posts, 9.1% of the total) offered a more radical and ideological perspective. It was the outlet that almost exclusively developed the narrative of an “institutional crisis,” linking the election results to a possible breakdown of the democratic system. Its content was reposted unaltered on Facebook and Instagram on multiple occasions, reinforcing a unified message of opposition to the rise of the right—which it labeled “fascist”—and denouncing alleged “U.S. interference” as a determining factor in the outcome.
Of the 187 posts from the Latin American block, 78 were devoted exclusively to narratives of conflict: 51 to fraud, 24 to foreign interference, and 3 to institutional crisis. The remaining 109 publications ranged from candidate profiles to coverage of the electoral process, but were framed within a context of polarization using terms such as “forced voting,” “fascism,” and “colonization.”
Russian block: Petro as fuel for technical suspicion
With 127 publications (34.1% of the sample), the group formed by Sputnik Mundo, Actualidad RT, and TASS was the second most active. Unlike the Latin American block, which focused much of its coverage on the confrontation between political actors, Russian media more frequently produced content on the technical aspects of the process. The initial allegations served as a starting point to explain how the electoral software works, the preliminary count, or to delve deeper into the debate over the procedures of the Colombian system.
Gustavo Petro’s statements took center stage in the coverage. At least ten publications directly used his allegations as a starting point to develop content related to fraud or foreign interference.
Two days after the first round of the presidential election, the president posted on X that he had “verified evidence of possible fraud.” Actualidad RT and Sputnik Mundo picked up the statement with similar headlines, without giving prominence to the institutional response or the reactions of election observers to these accusations.
Following Petro’s criticism of the United States and Israel, these media outlets also reinforced the narrative of foreign interference. For example:
- After Donald Trump’s public endorsement of Abelardo de la Espriella, Actualidad RT framed the Colombian president’s reaction as a response to a threat to sovereignty, with headlines such as “‘Freedom Dies’: Petro Reacts to Trump’s Endorsement.”
- That same approach continued on June 21, when Actualidad RT reported on Petro’s claims of an alleged hack of the pre-counting system attributed to Israel, reinforcing the idea of a democracy vulnerable to external manipulation.
After the presidential runoff, Petro retweeted a post by Actualidad RT on X that quoted Trump taking credit for De la Espriella’s victory. This was the only instance identified in the investigation in which a high-profile political figure explicitly amplified content from a Russian state-run media outlet.
In total, 55 publications of this block (43.3%) were related to fraud, foreign interference, or institutional crisis (36 on technical fraud, 18 on interference, and 1 on crisis). This demonstrates that the Russian media block not only repeated the allegation without providing any counterbalancing information—it turned it into a recurring narrative about the reliability of the Colombian electoral system.
Chinese Media Block: Institutional Framework and Interference
Media outlets from the People’s Republic of China accounted for 58 publications (15.6% of the sample), distributed among CGTN, Xinhua, and China Daily. Although their volume was significantly lower than that of the Latin American and Russian blocks, they also reacted to key moments in the Colombian electoral process and shared several of the predominant narratives.
CGTN (30 publications) was the most active and stood out for serving as a rapid source of content on X, providing real-time election data and results during voting days. Xinhua (with 27) provided more institutional and descriptive coverage, while China Daily had a marginal presence, with only one publication during the period.
Like the other outlets, this media group echoed the fraud allegations made by Gustavo Petro. However, the narrative that received the most attention was that related to foreign interference.
On June 21, the day of the presidential runoff, Xinhua introduced the concept of the “Donroe Doctrine”—a hybrid of Donald Trump and the Monroe Doctrine—to characterize Washington’s support for Abelardo de la Espriella as an act of “blatant interventionism”; all of this was presented in the format of an “interview with a political expert.”
Like the Russian media block, these outlets reported Petro’s statements on the issue from an institutional perspective, using headlines such as “Colombian President Presents Evidence of Alleged Fraud in the Recent Presidential Election” or “President Petro Rejects Donald Trump’s Direct Interference in Colombia’s Presidential Election”; thus, the criticism was entirely based on the Colombian president’s statements rather than representing the outlets’ own stance, and it failed to provide equal weight to the accounts of electoral institutions and authorities.
In summary, 21 of the 58 publications in this group echoed the doubts expressed by Petro: 11 regarding fraud and 10 regarding interference. Only 37 posts were strictly informative regarding the process. Their coverage reinforced narratives of interference and geopolitical factors—particularly those involving the United States—through academic concepts, interviews, and profiles to “disassociate” the editorial line from the opinions of third parties.
In addition to the coverage produced by CGTN and Xinhua, the People’s Republic of China also maintains content-sharing mechanisms with international media outlets. Among them is a partnership between the China Media Group (CMG) and teleSUR, the Venezuelan state-run channel with regional reach, according to the agreement renewed between the two countries in 2024.
How did this international coverage unfold?
The monitoring conducted between May 10 and June 23 shows a pattern marked by periods of low activity followed by sharp increases around key milestones in the electoral process.
1. The first round and allegations of fraud (May 31 to June 2): The conversation began to intensify on May 31, during the first round of the presidential election, with 47 publications focusing on the results and the prospect of a highly competitive runoff. Two days later, on June 2, there were 17 posts driven by Petro’s statements about possible fraud and Trump’s endorsement of De la Espriella.
2. The “Trump factor” (June 3–20): After that initial spike, the conversation entered a period of lower activity, with an average of nearly two posts per day about fraud for over 17 days. On June 10, Trump’s endorsement of De la Espriella reignited coverage with 17 posts—14 of them from teleSUR—and gradually shifted the focus toward allegations of foreign interference during the week leading up to the runoff election.
3. The runoff election as the main turning point (June 21 and 22): The highest volume was recorded between June 21 and 22, when the eight media outlets published 117 posts, accounting for nearly one-third of the entire sample. On election day, 60 articles were published, 29 of which were related to fraud or irregularities. The following day, coverage remained at virtually the same level (57 articles), although the focus shifted to the challenges filed by Iván Cepeda and Gustavo Petro’s allegations of an alleged “Israeli hack.”
It is worth noting that the similarity in the coverage by various media outlets corresponds to events and developments of high public interest—such as election days and statements by well-known political figures (Petro, Cepeda, Trump, etc.)—and is not evidence of transnational editorial coordination. Media outlets in Russia, China, and Latin America published content with nearly identical angles and headlines (e.g., Actualidad RT, Xinhua, Sputnik Mundo), relying precisely on quotes from statements issued by those involved in the process.
Upon applying a textual similarity analysis, 26 pairs of similar publications were identified; however, 25 of these 26 correspond to the same media outlet (teleSUR), which reused content across different platforms to generate wider reach, rather than indicating coordination among different media outlets.
What Was Left Out of the Coverage?
Although allegations of fraud did not make up the bulk of the coverage, the remaining 58.6% (218) of the publications also did not, in general, provide a comprehensive explanation of the electoral process. Rather, basic news content about the candidates, the campaign, and the results predominated.
However, the analysis reveals that this apparent neutrality was, in many cases, strategic: this content served to lend a veneer of journalistic rigor to the narratives of suspicion that dominated the agenda of the state-run media outlets monitored. During the second round, for example, the preliminary vote count was the main topic of 48 publications, while the National Registry appeared in only 5 and the National Electoral Council in just 3.
The National Registry is the clearest example of this dynamic. Although it was the most frequently mentioned institution in the study, with 31 references, its presence was primarily associated with questions about the transparency of the process rather than with explaining how it operates. In contrast, in media outlets such as teleSUR, Resumen Latinoamericano, Actualidad RT and Sputnik Mundo, mentions were concentrated between June 21 and 22, linking the preliminary count and the uploading of forms to an alleged “inevitability of fraud.”
This investigation does not demonstrate editorial coordination among the media outlets analyzed, nor does it claim to represent all international coverage of the Colombian elections. However, it does identify a common pattern: these outlets consistently devoted more coverage to moments of heightened political controversy than to contextual elements that would allow readers to understand or evaluate them.
Together with the national monitoring series conducted by Linterna Verde, this research shows that narratives about electoral fraud evolved and adapted to each of the major milestones.
In highly polarized electoral processes, this dynamic is particularly relevant. The international perception of an election depends not only on the events that occur on Election Day, but also on which of these events receive the most attention and which remain in the background. Understanding this balance is essential for analyzing how narratives and public perception of democratic processes are constructed in contexts of high political conflict.
***Methodological Note:
This research is based on a 44-day monitoring period, from May 10 to June 23, 2026, covering the first and second rounds of Colombia’s presidential election. The final corpus comprised 372 publications from eight international media outlets: teleSUR, Resumen Latinoamericano, Sputnik Mundo, Actualidad RT, TASS, CGTN, Xinhua, and China Daily. The selection was not based solely on a common editorial affinity but on two criteria: the direct ties of the media outlets included in the sample to the Chinese, Russian, and Venezuelan governments, and ProBox’s prior identification of these eight outlets as part of information operations targeting Latin American audiences. The objective was to analyze how this ecosystem covered the Colombian electoral process and which narratives gained the most visibility, without assuming coordination among these outlets or attributing a common agenda to them. This research is based on monitoring conducted over 44 days, from May 10 to June 23, 2026, a period covering the first and second rounds of the presidential election in Colombia.
The analysis included content published on X, Facebook, Instagram, YouTube, TikTok, and official websites. Data collection combined various techniques depending on the platform: BrandMentions, YouTube Data API v3, TikAPI, manual extraction, and scraping. Search queries were built using general terms related to the Colombian elections, the names of candidates and political figures such as Iván Cepeda, Abelardo de la Espriella, and Gustavo Petro, as well as concepts associated with the electoral debate, including “pre-count,” “fraud,” “interference,” and “challenge.” Each post was manually reviewed and classified. The sample was refined by excluding retweets, shared content with no editorial value of its own, and false positives identified through human review, with the aim of ensuring that the analysis focused solely on posts relevant to the Colombian electoral process.
Go to all articles
