The alliance identified the actors and industries behind online harassment that can often precede real-world violence.
Climate change misinformation often seeks to discredit science. In Latin America, it is increasingly targeting those who defend the environment.
To expose these attacks, the Latin American Center for Investigative Journalism(opens in new tab) (CLIP, for its initials in Spanish) and more than half a dozen partner news organizations launched an investigation in June 2026 to document online harassment against environmental defenders in six countries.
The project, “Digital Poachers(opens in new tab),” identified the main actors, narratives and patterns of intimidation targeting environmental defenders from Mexico through Panama and all the way to Brazil. The investigation found that the attacks mostly happened on X and Facebook, and that the mining and real estate industries were behind most of the harassment.
Las víctimas más recurrentes de los ataques digitales en Panamá son mujeres. Los casos de @raisabanfield @cgadames @martitacornejor revelan cómo funcionan las operaciones para desacreditar y hostigar en redes sociales #TaladoresDigitales @CLIP_AmericaLat https://t.co/HaroFbjg3w
— Concolón Panamá (@ConcolonPanama) June 16, 2026
“The disinformation machine says, ‘These people don't know what they're talking about, or they're being paid, or they actually have other interests,’” Pablo Medina, CLIP's misinformation editor, told LatAm Journalism Review (LJR). “Many of those claims are misinformation.”
“Digital Poachers” developed a collaborative methodology for investigating these misinformation campaigns. The combination of collaborative reporting, a shared database, narrative analysis and AI-powered automated workflows offers a roadmap that other newsrooms and academic teams can adapt to document and investigate similar cases.
CLIP brought together news organizations with experience covering environmental issues in the Latin American countries that are considered the most dangerous for environmental defenders. It partnered with AFP Checamos(opens in new tab) (Brazil), ColombiaCheck(opens in new tab) (Colombia), Agencia Ocote(opens in new tab) (Guatemala), Contracorriente(opens in new tab) (Honduras), Concolón(opens in new tab) (Panama), Animal Político(opens in new tab) and Mongabay Latam(opens in new tab) (Mexico). The project was supported by UNESCO's Global Initiative for Information Integrity on Climate Change.
CLIP designed a database and shared it with all participating newsrooms. Each partner tracked digital attacks in their countries, added them to the database and adjusted the structure as they went along, Medina said.
To document the attacks, the newsrooms interviewed organizations and environmental activists, searched social media, and used software to identify users and inauthentic content whose main goal appeared to be to manipulate a narrative.
After nearly two months of data collection, CLIP standardized the information provided by its partners and built a database containing nearly 800 posts documenting digital attacks against individuals, communities and organizations defending the environment between 2017 and 2026.
Using the database, CLIP created a series of interactive elements that allow users to browse and filter the information by type of attack, platforms used, types of victims (individuals, Indigenous communities, civil society organizations or journalists) and the industries to which the alleged perpetrators belong, among other criteria.
The project also includes nine stories(opens in new tab) of attacks and harassment targeting environmental defenders in the countries where the participating media outlets are based.
Àngela Cantador (L) was part of the CLIP team responsible for data processing, while Pablo Uribe (R) served as the project's editor. (Photo: Courtesy CLIP)
It is an “attempt to understand how these digital attacks happen and how they can later have repercussions in the real world,” Àngela Cantador, CLIP's data coordinator, told LJR. “It was more an exercise in evaluating these narratives to give other journalists and academics clues about how they are generated and how they can be documented and analyzed.”
Unlike other CLIP investigations in which the databases are made available to the public, the data from “Digital Poachers” was not released. The team said this was to avoid revictimizing those targeted by the attacks and amplifying the disinformation.
“This was the first major challenge: How to handle this data so that we could show our findings without exposing people,” Cantador said. “That's where the idea of analyzing the narratives came from.”
CLIP and its partner news organizations initially agreed to classify the attacks into five categories, based on the cases they had encountered most often in the past: reputational attacks, gender-based attacks, racist attacks, climate misinformation and “greenwashing(opens in new tab),” or misleading information about certain industries’ environmental practices.
CLIP's tech team developed an automated workflow in Python that used OpenAI tools to analyze the attacks and propose four new categories.
Based on that input, the team subdivided the “reputational attacks” category, which contained the largest number of entries, into criminalization, stigmatization, corruption or hidden interests and personal attacks.
With the new classification system in place, the team ran the data through the AI model again and manually verified the results at least four more times before building the final database, Cantador said.
“AI can be really useful in cases like this to help you work with large volumes of data, come up with an initial classification or identify inconsistencies, but there always has to be manual verification,” Cantador said.
More and more news organizations are leaning on AI-powered automated workflows(opens in new tab) to aid them in large data-driven investigations. Many are using open-source platforms such as n8n(opens in new tab), which are more visual and intuitive and do not require extensive coding knowledge.
However, the programmers on CLIP's technology team chose to build their own automated workflow using a Python script for the “Digital Poachers” project.
“Running an automated workflow to analyze large volumes of data is, I think, very important in journalism right now, because there is so much information that processing it manually is very difficult,” Sairo Alemán, a front-end engineer at CLIP who handled the project's technological development, told LJR.
#TaladoresDigitales Defender un río le costó a Reynaldo Domínguez el exilio, la persecución y la pérdida de sus hermanos. Durante años, él y su comunidad resistieron al avance de un proyecto minero en Guapinol. Hoy vive lejos de su tierra, pero sigue recordando a quienes fueron… pic.twitter.com/ZjZZTYZZ9C
— Contracorriente (@ContraC_HN) June 11, 2026
The team behind “Digital Poachers” said the project is intended to be a starting point for further journalistic and academic research into online attacks and environmental misinformation.
“It is very important to make people understand that we are all victims of misinformation, but in these cases, there are victims who may actually end up dead,” Jennifer Ávila, co-founder and editorial director of Contracorriente, told LJR. “This goes beyond even the right to truth or to be informed; it is also about the right to life.”
Translated with AI assistance and edited by Alejandro Martínez-Cabrera
