Nearly one-in-four unique accounts that follow or amplify posts by the prominent anti-Israel profile @Wallface exhibit strong bot-like signals, according to a study released by the Antisemitism Research Center (ARC) of the Combat Antisemitism Movement (CAM) on Wednesday.
The study, titled “Bots, Fake Profiles, and Inauthentic Accounts Amplifying @Wallface on X,” was conducted between October 19 and October 20, and examined 10,709 unique X/Twitter accounts interacting with the profile.
The account @Wallface has roughly 51,000 followers on X; the account bio says: “Research and data analysis” and shares content critical of Israel.
The research forms part of CAM’s Bots Accountability Project, an initiative that documents and reports bots, fake profiles, and inauthentic accounts while tracking platform responses.
To evaluate the accounts, ARC used CAM’s detection model, known as X model v1.4.0, which assigns every account a score from 0 to 100 based on signal weights grounded in research from the Digital Forensic Research Lab, Indiana University’s Observatory on Social Media, and Cresci studies.
Twenty-five percent of accounts show signs of being bots
The methodology evaluates signals including account age, posting rates and activity bursts, follower ratios, handle patterns, default avatars, and profile completeness, while ensuring that post content and ideology are not used as inputs.
The report applies a conservative cutoff score of 75 or higher, meaning that each step up removes accounts that still look automated, ensuring the figures understate rather than overstate the presence of inauthentic profiles.
Out of the 10,709 unique accounts analyzed in the sample, CAM’s model identified 2,597 accounts, or 24.3%, as showing strong signals of being bots, fake profiles, or inauthentic accounts.
The breakdown across specific user groups showed that 27.2% of the newest followers sampled, 23.7% of reposters, 13.9% of commenters, and 8% of quoters met the threshold for bot-like activity.
The study found that the proportion of flagged accounts among amplifiers grew as the profile’s posting frequency accelerated over the months. Among accounts amplifying posts published in July, 15.9% showed automated signals, climbing to 19.1% in August, reaching 23.2% in September, and settling at 22.2% in October, during which time the account’s daily output jumped significantly.
The analysis highlighted that certain individual posts attracted exceptionally high concentrations of automated engagement.
Heavy volume of automated accounts
An October 16 post asserting that Israel maintains a paper trail for its social media operations was amplified by 86 accounts, of which 31.4% were flagged as inauthentic. Similarly, an October 19 post in which @Wallface complained about being targeted by a coordinated bot-reporting campaign faced a heavy volume of automated accounts, with nearly a third of its 68 amplifying profiles showing clear signs of inauthenticity.
Further investigation into the flagged profiles revealed significant overlap with previous research. Out of the thousands of automated profiles detected, 428 had previously surfaced in past CAM investigations, with the vast majority receiving similar inauthentic ratings.
The most prominent overlaps occurred with accounts previously identified around figures such as Roger Waters, where 237 shared profiles appeared, and Jake Shields, with 162 overlapping accounts, alongside smaller overlaps linked to networks surrounding Ana Kasparian, Abdul El-Sayed, Rabbi Jeffx, and Mario Nawfal.
Among the specific examples highlighted in the data, a profile named @Rajdhani108306 achieved a near-maximum bot score of 99.9 out of 100 after generating roughly 171,000 posts since its creation in June, translating to an average output of more than 1,200 posts every single day.
CAM noted that the results “raise concerns about how bot-like amplification can distort perceptions of online influence and public support. The recurrence of flagged accounts across multiple audiences suggests the problem extends beyond any single X account.”