On August 20, three Facebook users – Shani Assaf, Uri Ya’ar and Julia Friedman – all decided to change their names and profile pictures. These changes were no coincidence. The next day, we revealed that they were part of a group of fake profiles operated by an influencer agency called The Leaders. Full information about the group was sent to Facebook, yet those profiles remain active today, as is the norm on social media.
The network consists of around 25 fake profiles. A review conducted last week found that they have been changed to include new pictures and new usernames. For instance, Julia Friedman has become Julia Freeman, Shani Assaf has become Shani Assif and Uri Ya’ar, a guy with a beard and black hair whose face was hidden behind a camera, has become Uri Ya’ari, a blond who smiles straight at the photographer.
Even though this is a violation of Facebook’s community standards, the company is still letting these fake accounts continue their manipulative activity of responding to comments and sparking discussions for the sake of disseminating messages on behalf of people or companies with specific interests, commercial or otherwise. Even the Consumer Protection and Fair Trade Authority, which argues that fake responses, recommendations and criticisms necessarily mislead consumers, hasn’t done anything to stop this practice, claiming it lacks the necessary resources.
The Leaders is just one of dozens of agencies working to promote the interests and messages of commercial companies. The service they offer is called “monitoring and intervention.”
The agencies begin by monitoring Facebook discussions relevant to the brands that hired them. In the next stage, they deploy real users who ostensibly tried the products in question to try to sway the discussion in favor of their customer’s brand, or alternatively to criticize a rival brand.
The agencies also use fake profiles that they created specifically to look like real responders to spark discussions that will convey the messages their clients want sent. They thereby artificially manipulate consumers who joined Facebook groups to get advice from real users who could share information gained by personal experience.
The Leaders is still operating the group of profiles uncovered in TheMarker’s investigation, but under new identities. One way to prove that the profiles in question are avatars – fakes fully operated by a real person who uses them to influence the public conversation – is to look for their profile picture on Google’s search engine, which will bring up the original from which it was taken. TheMarker’s initial investigation found that all the profile pictures came from free online photograph databases.
- Anti-Netanyahu activists say this false flag campaign is an attempt to kill their protest
- Inside the secret Israeli industry of fake Facebook profiles
- Facebook removes fake account posing as protester comparing Netanyahu to Hitler
Now, however, the system has been improved, and the new fake profiles are based on what is known as deepfake technology. In other words, the profile pictures were apparently taken from a website that uses artificial intelligence to create pictures of people who don’t exist. One example of such a site is This Person Does Not Exist, which, with the touch of a button, can create a head shot that looks like a real person.
“Anyone who uses Facebook today needs to know that he is in a universe of fakes on top of fakes,” said Dr. Tehilla Shwartz Altshuler, a senior fellow at the Israel Democracy Institute. “The first fake is the way Facebook itself organizes our feed, in a way that reflects not objective reality, but Facebook’s commercial interest in having us waste more time on the web. The second fake is all the millions of organizations that live off their attempt to manipulate Facebook’s algorithms – the fakes, the avatars and the influencer agencies. Anyone who uses social media ought to understand the double fake that it exists in.”
‘Authenticity is the cornerstone of our community’
In Facebook’s latest transparency report, the company said that during the first half of 2020, it erased 3.2 billion fake accounts. Facebook claims to have 2.7 billion active users per month.
The company even claimed to prioritize enforcement against fake accounts. “Authenticity is the cornerstone of our community,” it said. “That's why we require people to connect on Facebook using the name they go by in everyday life.”
It added that it works hard to limit the spread of spam, because it doesn’t want to allow content that is meant to deceive or mislead users simply for the sake of increasing viewership.
Nevertheless, many fake accounts continue to exist on social media, for one simple reason – they serve the companies’ business model. In other words, the virality that fake profiles create through their social media activity also enables the company to earn huge profits from advertising.
Dr. Tomer Shadmy, an expert in law, ethics and technology at Hebrew University’s Federmann Cyber Security Research Center, said that despite the social media companies’ claims to be working against the practice, the data proves otherwise, since it shows that the problem isn’t decreasing.
Social networks, and especially Facebook, could take action against fake profiles if they wanted to, she argued, because only they have information about who is behind those profiles. They also have the technological capability to identify patterns of behavior typical of such accounts when they are active.
“At a time when the companies are in the best position to identify fake accounts, the social networks’ business model and interests only incentivize such accounts,” Shadmy said. “The companies want to show investors that the number of users is growing constantly, and those accounts are part of their user base. Moreover, fake accounts create more interactions – clicks and responses – which the companies are constantly trying to maximize as part of their financial model.”
“As long as a fake profile exists and is posting content that creates traffic jams for Facebook, they can insert more ads,” added another researcher in the field of fake news and manipulation by social media users.
Facebook, which earns 98.5 percent of its income from ads, is apparently an extremely effective vehicle for selling advertising space. Its earnings have increased regularly from year to year.
An avatar for everyone
The most famous Israeli case of a fake profile that sought to create a false impression was an account named Dana Ron that posted a threat against Prime Minister Benjamin Netanyahu. “Bibi has to be taken down by force … dictators are removed only by a bullet to the head!” it said.
Yossi Dorfman, a strategic consultant on digital affairs, proved in a series of tweets that the account was fake, and Facebook removed it that very same day. An investigation by the police’s cyber division discovered that the account was run by a foreign resident posing as a woman named Dana Ron. But police didn’t say how they dealt with the person who operated the account.
It’s important to recognize the differences between different types of fake profiles. “A bot is a profile that impersonates a real person but is run by computer commands,” Dorfman said. “An avatar is a character fully operated by a real person who uses it to influence the conversation. Overseas, it’s common to use the term ‘sock puppet account’ to describe this kind of activity.”
According to Shadmy, we shouldn’t rely on self-regulation by giant internet companies to locate and remove fake accounts.
“Intervention in this field is essential,” she said. “In Israel, the law already allows prosecuting someone who manages and uses fake accounts to deceive people about commercial and consumer affairs. The use of real pictures and other details on fake accounts is also a crime. One important piece of future legislation could be forbidding people to operate or use the services of companies that offer influence operations via fake accounts.”
The exposure of some of the avatars apparently operated by The Leaders a few months ago happened after Teva Pharmaceuticals filed a 40 million shekel ($12 million) lawsuit against the agency and Abbott. Teva claimed The Leaders had used fake accounts to promote Abbott’s baby formula, Similac, by smearing Teva’s rival brand, Nutrilon.
A month ago, attorneys Liad Vertzhaizer, Ohad Rosen and Hagai Kalai filed a class-action suit in the Tel Aviv District court over the same issue. “The Consumer Protection Law explicitly prohibits advertising ‘that is liable to lead a reasonable man to assume that what is said in it isn’t an advertisement,’” the suit said. “The time has come for the Consumer Protection Authority to use the very substantial tools given to it to start seriously enforcing the prohibitions in this law.”
Shwartz Altshuler said that responsibility for preventing such practices rests with the Consumer Protection and Fair Trade Authority rather than Facebook itself. “The authority needs to generate enforcement against anyone who tries to mislead consumers, and it’s the one that should, through Facebook, expose the groups behind the fakes. It’s convenient to point an accusing finger at Facebook when the problem lies with the local enforcement agencies.”
"Ads on highways with heavy traffic are worth more because more people will see the billboard for longer as they sit in traffic. This logic holds just as true for Facebook. Every fake profile that is active and posts content online helps create a ‘traffic jam’ for Facebook where they can place more ads,” said another researcher who requested to remain anonymous but whose work focuses on manipulation of social media.
Facebook said its global task force had examined the profiles, and any that showed signs of being inauthentic were removed from its platform.
The Leaders said, “This information is recycled and rehashed. The Leaders is a company that specializes in building brand communities and in marketing based on experience, a company that knows how to turn a satisfied customer into an experienced ambassador, and it doesn’t operate through fake profiles.”