For any user of social media this scene must be familiar: with the idea of seeing someone's profile - a singer, a politician, a content creator - we type their name in the search engine and after hitting enter the screen fills up with accounts with images of those people, with names similar or identical to the one we want to find. Suddenly, the search seems like a game of concentration: which is the real profile and which is fake? Sometimes, the blue verification pimp jumps out to clear our doubts. At other times, it takes a keener eye to find the person we are looking for.
The platforms have policies to sanction anyone who impersonates someone else or creates pages to falsely represent a brand or organization. Of course, these kinds of practices can be associated with crimes. Among other things that the impersonation rules aim to prevent are scams, misrepresentation of political views - when someone appropriates the identity of a candidate, for example - or tackling sex offenders who use another name or appearance to approach their victims.
You don't need to have thousands of followers or be a public figure to fall into the hands of impersonators. In December of last year, CNBC reported that Bob Kurkjan, a U.S. military man, had found nearly 40 Instagram accounts with his photos and minimal variations of his name. From the descriptions, it was clear that these digital clones were being used for scams. The situation, already a bit disturbing, worsened one day when Kurkjan wanted to log in to his account and discovered that it had been blocked. To get it back, she had to send an image of her passport for Instagram to verify her identity.
When we discover that someone is impersonating someone else or ourselves, we can follow the process provided by the platforms to report the accounts, which is usually found in the three-dot menu on the profiles. Unlike other rules, such as nudity or hate speech, where violations are detected by automated systems, impersonation relies more on user reporting. But this does not mean that sharing the account for others to report - as so often happens - serves to expedite the review process of the reported profile.
In any case, creating an account with someone else's name or photos is not always prohibited. On Twitter, for example, parody accounts, accounts for publishing news about someone, or fan accounts, which most often use similar names and images to report or to mock or support a public figure, are allowed. To avoid penalties for impersonation, administrators must indicate in the biography and account name - which is not the username - that there is no link to the real person. To make this distinction clear, the company suggests using formulas such as "fake", "fan" or "comment".
Facebook -which differs from Instagram in this respect- wants its users to identify themselves on the platform in the same way they would in real life. That is why its policies require people to use the name by which they are usually called and that appears on an identity document, although it allows nicknames if they are a variation of the real name. These rules -which apply to profiles, not pages or groups- are so strict that they even limit those people who want to display their diplomas in their Facebook name, the case of those who register as "Engineer Pedro Pérez" or "Doctor Juan González", as it is forbidden to include any kind of title. Failure to comply with these policies could result in temporary restrictions or permanent suspensions.
These policies have been highly questioned, as they directly affect transgender people or drag queens, who would have to register with the name that appears on their documents even if they do not identify with it. The problem also affects, among others, people who for security reasons must protect their identity, and who would be exposed under these rules.
In any case, the common purpose of impersonation rules is to avoid deception and prevent criminal behavior. It is often believed that account verification (the blue pimp) can help mitigate this problem -in fact, platforms present it as a way to guarantee the authenticity of a public figure or brand-, but rather they have become a status symbol within the networks, as well as a focus of criticism for the platforms.
In 2017, Twitter had to suspend verification processes after awarding the badge to Jason Kessler, organizer of the far-right Unite the Right rally. A year earlier, the platform also came under fire for verifying the account of Milo Tiannopoulos, a blogger known for his extremist positions. Twitter ended up removing his blue pimp for violating its rules, but that revocation - as suggested by specialist journalist Casey Newton - could indicate that verification was more than just a measure to ensure the authenticity of an account.
Beyond that, the important thing is to differentiate a measure that confirms the authenticity of an account from the procedures to combat impersonation, which, as we saw, falls on users to report it. According to Twitter figures, in the first half of 2021 about 200,000 accounts were suspended for this reason. Facebook, for its part, reported that between July and September 2021 action was taken against 1.8 billion fake accounts, a category that includes both cases of impersonation and profiles created to manage coordinated harmful activities.