Anonymous blogs got Google to remove from its search engine press articles about businessmen, politicians, and swindlers by falsely alleging plagiarism.
Por Marco Dalla Stella y el CLIP
If you google the name Edgar Manuel Méndez Montoya, financial trader of former governor of the Mexican state of Quintana Roo Roberto Borge, you will find a strange disclaimer among the results: “In response to multiple complaints we received under the US Digital Millennium Copyright Act, we have removed 4 results from this page”.
In other words, Google reports having received several complaints alleging that the removed articles were the product of plagiarism, and as such, violate U.S. copyright law (Digital Millennium Copyright Act, or DMCA).
One of the deleted articles, “Warrant issued for the arrest of Edgar Méndez Montoya, alleged trader for Borge”, was published by XEVT, a local radio station in the state of Tabasco that won the National Journalism Award in 2013. The article reports an arrest warrant issued against the Tabasco businessman, accused of embezzlement. Its deletion came to light after reviewing the Lumen database, the project of the Berkman Klein Center for Internet and Society of Harvard University, which registers all removal requests -for any reasons- received by Google, among which alleged plagiarism.
Google agreed to remove the Tabasco radio article after a person identifying herself as Lucía Montes Celester, claimed to be the intellectual owner of that content. She stated under oath that radio XEVT had copied their content from anonymous blog newsworldinfo.over-blog.com. All she had to do was to prove that her blog entry had been published before the article came out.
However, no identification is needed to submit this kind of claim, and people are free to give their blog entries any date they want. There is no filter to prevent people from using false credentials and claiming plagiarism. Lumen is rife with requests for removal of content made under false names, such as Adolf Hitler, Joseph Stalin, and Elon Musk. In response to a query sent by CLIP, Google said it is in fact unable to verify the accuracy of these claims.
Following the trail left by this censored article about Méndez Montoya, the Latin American Center for Investigative Journalism (CLIP), reviewed copyright infringement complaints for other articles published by newsworldinfo.over-blog, where the story from the radio had appeared.
We found that Google received seventy other requests for removal from the same blog, all between late 2022 and 2023. The articles in question contained damning reports about several big businessmen and people with political connections which stood to gain from hiding news about their past. Moreover, the people in question where not only Mexican, but from other countries in the region such as Venezuela, Paraguay, and Haiti, and from other parts of the world, such as Italy, Canada, or the United States.
Using as a starting point the articles copied in the blog just mentioned, we searched Lumen’s database, using a programming code, for other anonymous blogs that might have been copying content and then sending removal requests to Google. We found about 30,000 interrelated removal requests. That is, either the webpage that issued the request was the same, or the webpage being attacked was.
More than 14,000 of these removal requests originated from anonymous web pages registered on blog platforms such as Blogspot, Tumblr, and WordPress. Almost all of them date back to 2021, and so far, more than 3,000 requests have been made in 2023.
This investigation further found that these thousands of removal requests are linked to a common source, with connections to Spanish digital reputation company Eliminalia.
As an international investigation led by Forbidden Stories revealed last February, Eliminalia was in the business of getting platforms to delete compromising customer information. Through leaked internal documents of the company, this journalistic alliance revealed a list of renowned clients who in recent years had hired Eliminalia or its allied companies. Last December, just before this was revealed, the company changed its name to iData Protection.
What is interesting is that the database collected and analyzed by CLIP partially corresponds to the list of clients released in the Forbidden Stories-led investigation. To cite just one example, some of the requests we reviewed were aimed at hiding critical news about Venezuelan businessman Isaac Sultan, who, according to Forbidden Stories, paid $100,000 to hide information about his network of companies.
This CLIP investigation unravels for the first time the workings of a fraudulent method for hiding shady pasts through removal requests from anonymous blogs and identifies cases in which Google fell into as simple a trap as changing the date of a piece of content on an anonymous blog. In addition, based on the targeted articles, we were able to put together a list of more than fifty names of individuals or companies that could benefit from such removal, which allows us to assume that iData Protection – the new name under which Eliminalia still operates – or other similar companies have continued operations over the last three years. Of the removal attempts reviewed by CLIP, more than 3,000 took place after the alliance led by Forbidden Stories made its revelations.
According to the pattern we identified, consulting firms create anonymous blogs, including some on Google-owned platforms, where they copy the articles that their clients want removed. After changing the date to a time before the original publication, they claim copyright infringement to get the articles removed.
This investigation has found that over the past three years, Google has removed at least five hundred pages due to copyright claims that appear to be fraudulent. Today, it is still impossible to access many of these pieces through the search engine.
Regarding these findings, a Google spokesperson answered to CLIP’s questions by email, assuring that they were aware of this practice and admitting that a percentage of the removal requests have slid through their security checks. He also said that they are in the process of re-including links that have been erroneously taken down.
While the owners of the censored articles can appeal the decision, several authors contacted by CLIP say they have not received any communication regarding the removal, and therefore their articles are still down.
“We know there is no proper notification,” said Tord Lundström, technical director of computer forensics at Qurium, a company that provides hosting services for independent and investigative newspapers, including CLIP.
Qurium was among the first organizations to raise the alarm regarding fraudulent DMCAs. According to Lundström, newspapers hosted by Qurium have been subject to several fraudulent DMCA attacks, but Google never notified them. “Because we are a hosting service, we are very easy to contact,” Lundström said.
The text of the copyright protection law, the Digital Millennium Copyright Act, demands that service providers notify the person whose content has been taken down.
However, Google claims that unless they are previously registered in Search Console, a Google service that allows web page creators to check the status of their sites on the Internet, they do not have the means to notify all users. Without Search Console registration, the platform claims that it cannot identify the owner of a web page.
This flaw in the system allowed digital consulting firms to use the creation of removal requests for copyright infringement as a tactic. Under the pretense of protecting people’s right to be forgotten, they try to clean up the reputation of indicted politicians, white-collar criminals, and fraudsters.
The right to be forgotten
The business of digital reputation is as old as the internet. In the beginning, the only way for individuals to improve their online image was to turn to companies specializing in search engine optimization, or SEO.
Things changed in 2014, when the Court of Justice of the European Union made a ruling on a legal battle between a Spanish citizen, the Spanish Data Protection Agency, and Google. At that time, the Court of Justice decreed what is now known as the “right to be forgotten,” which consists of the right of individuals to obtain the removal of obsolete or irrelevant information from search engine results. Excluded from this decision were individuals whose past actions were of public interest, such as politicians and businessmen.
The Court’s decision created a legal framework for the emergence of new companies and law firms legitimately specialized in the right to be forgotten. Standing on muddier ground, a parallel business grew of selling to politicians and businessmen all over the world the possibility of hiding their bad steps and cleaning up their public image, although the ruling of the European high court did not protect them.
“We were no longer applying the law as we were supposedly hired to do,” a lawyer specialized in the right to be forgotten who worked at Eliminalia told CLIP. He asked for his name to be left out because he fears reprisals, a sentiment shared by other sources consulted for this investigation.
The lawyer explained that his job was to convince Google and other search engines that they had to remove the content about their clients, using any method of appealing to the right to be forgotten created by the European Court. “Practically, it was about looking for ways of tricking Google”.
But the clients he started getting didn’t have strong legal cases. “I started getting clients who were politicians, drug dealers, rapists,” the lawyer said. His job became even more difficult in 2019, when the European Court of Justice ruled that the right to be forgotten could not be applied for people outside the European Union.
“That’s when we got screwed,” the lawyer said.
Clients paid the company €150 to €3,000 for each link it managed to hide from the top results on the search engine, depending on how much public interest the article had aroused. If Eliminalia managed to drag a link to an article to the fourth page of Google results, it considered it deindexed, the lawyer said. When the lawyers’ claims of content removal were unsuccessful, the company would appeal to its plan B: try to hide negative content by creating positive content.
Positive pieces
Instead of turning down clients who did not have valid claims to the right to be forgotten, companies like Eliminalia reverted to the old trick of hiding unwanted results by creating lots of similar-looking content. As CLIP was able to establish, companies dedicated to erasing the past of dubious clients have entire departments dedicated to producing new articles that distort the content of the original, until the targeted one drops down to the second or third page of Google search results.
“I would spend eight hours writing articles, one every hour or hour and a half. From 1,500 to 5,000 words per hour. You’re like a machine,” a former employee in the content creation department of another smaller digital reputation company told CLIP.
“The goal is to manipulate information,” the former employee told us. She explained her job was to create fake narratives in series, publish them on websites with no real audience but which, because of her knowledge of SEO (search engine optimization), would be well ranked by Google’s algorithms.
“What you’re writing about a person doesn’t really exist,” she said.
Sometimes, to spread the information, she said the company would ask her to create fake profiles on all the social media.
“We had sim cards that we changed every few days to receive activation codes. I don’t even know where those cards came from. We would change like two hundred cards a week,” the former employee said, explaining that fake profiles were created on all major social media platforms such as Facebook, Twitter, Pinterest, and Instagram.
According to this former employee, workers were closely surveilled by the company, as many of the clients were politicians and other men with power.
“We couldn’t use cell phones, and there were a lot of cameras,” the former employee in the content creation department told us. “It’s all under surveillance.”
The former employee decided to leave after realizing the gravity of what she was doing. Since she was not allowed to know the reasons why clients had approached the company, she once decided to Google the name of a client she had been writing many articles about. She found many accusations of abuse, including towards minors. “They tell you not to believe everything you find on the Internet, so that when you have to write you don’t feel apprehensive, so to speak,” said the former employee. Her perception of the world changed radically after this experience.
“You realize that we are never safe from anything. Because there are people ready to erase you from the face of the Earth,” the source said.
Copyright in the digital age
When content creation proved insufficient to bury a thick past, companies like Eliminalia resorted to abusing the 1998 U.S. law that sought to limit the dissemination of pirated materials on the Internet.
As another former Eliminalia employee told CLIP, this tactic became popular after 2019 when the European Union introduced limits on the application of the right to be forgotten, excluding people outside Europe.
“What we were doing was using websites to create copies of the content we wanted to delete,” he said. “You’d put a date on it that was prior to the original publication, and Google would delete the original because it seemed to have been copied.” According to this second former employee, the tactic proved particularly effective at first if the copied content was uploaded to blogs created on Google-owned platforms.
“I would say sixty percent effective if the content was created on websites. And one hundred percent if it was created on Blogspot.”
This tactic takes advantage of the fact that users of blogs and similar platforms can alter the publication date of their content, apparently without leaving any trace. “If Google didn’t provide this possibility, that really couldn’t be done”, the former employee said.
The combination of abuses of U.S. copyright law and European regulations on the right to be forgotten worries Catalina Moreno Arocha, Coordinator of the Social Inclusion Line of Fundación Karisma, a civil society organization promoting human rights in the digital world. “This can become very dangerous,” she said.
In some Latin American countries there are laws and data protection authorities that apply a right similar to the right to be forgotten, but there is no such figure as the right to be forgotten as defined in Europe in the region at large. However, the Special Rapporteur for Freedom of Expression of the IACHR has said that this right is “problematic” in the context of the Inter-American Human Rights System. Despite this, there are companies offering to erase your past if you can pay.
“It sounds good to talk about a ‘right to be forgotten’, but it’s imprecise,” Catalina said. “There is no such right. It’s a form of censorship, a threat to freedom of expression.”
When asked what instruments are available to a citizen affected by news damaging to his or her reputation, Catalina points to existing protections against libel and false information, and the possibility of having an article updated. For all other cases, the principle of freedom of expression comes first.
“The information has to stand,” she said.
How we investigate in detail
As explained above, CLIP’s investigation started from a Tabasco radio article that was removed from the search engine at the request of newsworldinfo.over-blog.
We then checked the blog for other copied material. We found that at least sixty more requests for removal had been issued to Google from this blog, all between late 2022 and the present. To streamline this series of interrelated searches into the Lumen database, we developed a code.
We then looked for other anonymous blogs that were copying content and sending removal requests to Google. For example, the same article about a massacre in Haiti copied by newsworldinfo.over-blog.com was also copied by the blog https://mundololopez.wordpress.com. Another one about the arrest in Spain of two former Venezuelan officials was copied by https://theobservernews.over-blog.com/, and also by https://mundololopez.wordpress.com.
By automating this search process, we obtained a database of about 30,000 interrelated removal requests. That is, they have in common the web page from which the request originated or the page that was being attacked.
In most cases we found that removal attempts were aimed at deleting negative information about forex trading platforms. Forex trading is a strategy for profiting from the foreign exchange market.
According to the U.S. Commodity Futures Trading Commission (CFTC), a U.S. government agency that regulates futures and options markets, fraud is common in the forex trading market and has increased with the spread of technology.
After excluding orders related to forex trading agencies, we were left with a set of nearly 5,000 requests for removal.
As a last step, we compared those five thousand removal requests with the Google Transparency Report database and found that Google accepted complaints from anonymous blogs, mostly registered on Blogspot and Tumblr, and removed links for more than three hundred pieces, for alleged copyright infringement.
It is important to mention that removal requests that did not come from blogs or their authors were excluded from our calculation. Sometimes, actors engaged in submitting claims of U.S. copyright law violation pose as the authors of legitimate articles and thus delete compromising information for their clients. For example, Lumen shows more than fifty removal requests made from the blog infodio.com, owned by Venezuelan investigative journalist Alek Boyd. Using his or his blog’s name, someone else requested the removal of articles in other media platforms. “All requests for removal of information related to my blogs have been made by anonymous or false actors, and in some cases they have used my own name and that of infodio to remove infodio information republished on other blogs!”, Boyd told us.
CLIP did not focus on cases such as Boyd’s, but only on those where the claimants of alleged plagiarism were anonymous blogs or publications. For this reason, it is likely that the numbers mentioned in this article are actually much higher.
To find out who is behind the blogs sending the requests, we shared the list of blogs registered on Tumblr and WordPress with Automaticc, the company that owns both those platforms. Unfortunately, an Automaticc spokesperson told us that they could not share information regarding its domains, and assured us that they started suspending pages that violate their policies.
“We recognize that there is a lot of potential for abuse, especially as a way to censor speech and legitimate criticism,” Automatic’s spokesperson told CLIP. “We believe that laws about online speech like the DMCA should not compel hosts to aggressively remove content and enable this type possible.”
Since hosting services do not have to disclose the names of their blogs’ owners, CLIP’s investigation could not name with certainty the individuals or companies behind them. However, in the database that resulted from this investigation we found matches to people or entities that had used Eliminalia’s adverse content removal services, as reported by the journalism consortium led by Forbidden Stories.
In an alarming number of cases, removed pages are from well-known newspapers and mention people that European law would likely consider public figures. For example, Google deleted from its results a page from Offshore Leaks, the portal of the International Consortium of Investigative Journalism (ICIJ), due to a copyright infringement notice sent by tumblr.com/eryanrealty, a page that is currently not even online.
In written communications, Google’s spokesperson assured CLIP reporters that the company actively fights against fraudulent removal attempts, employing a combination of human and automated review. “We also consider the significance of the content requested to be removed, and take extra care in appropriate cases to avoid wrongly limiting access to information,” the spokesperson said.
This story was produced with the support of the International Center for Journalists’ Disarming Disinformation program, a three-year global effort with lead funding from the Scripps Howard Fund.
