AI Bot Networks and Foreign Interference Target Recent Israeli Elections
Reports reveal foreign influence and AI-powered bot networks flooded recent Israeli elections with fake accounts to manipulate public opinion, promote candidates, and undermine trust.
A massive covert influence campaign involving AI-generated bot networks has targeted recent Israeli elections, flooding social media with fake accounts designed to promote specific candidates, attack political rivals, and undermine public trust in state institutions.
According to an investigation by Scoper conducted in the months leading up to the vote, significant anomalies were uncovered in the user profiles sharing content from major party leaders. The data revealed that 71.5% of the accounts promoting Prime Minister Benjamin Netanyahu were identified as non-Israeli, with over a third featuring random number sequences in their handles. Furthermore, a sample check of approximately 194,000 new followers added to Netanyahu's accounts showed that 75% had been created just 24 hours prior to the inspection. By comparison, non-Israeli shares for opposition figures stood at 12.4% for Yair Golan and 10% for Gadi Eisenkot.
The coordinated manipulation also targeted smaller and newly formed factions. The Fake Reporter organization identified roughly 100 fraudulent accounts operating against Ofer Winter and the Amcha Yisrael party just a day after its establishment. The perpetrators utilized AI-generated profile pictures and disseminated hundreds of tweets within an hour. Concurrently, the Yashar party, led by Eisenkot, lodged complaints regarding the distribution of a forged official statement bearing its name. Additional networks were found promoting Itamar Ben Gvir's Otzma Yehudit and Avi Maoz's Noam party, though both political entities categorically denied any connection to the activity.
External interference was additionally documented by the Institute for National Security Studies (INSS), which exposed the "Isnad" network. This operation ran approximately 100 accounts masquerading as Israeli citizens, churning out 25,000 posts aimed at stoking societal divisions. Social media platforms responded with mass purges; TikTok removed over 45,000 videos and 1,400 accounts for violating electoral regulations, while law enforcement agencies dismantled dozens of suspicious profiles on Twitter and Facebook days before polling stations opened.
Chen Barakat, a technological entrepreneur and artificial intelligence expert, explained that modern tools enable cheap and rapid influence operations. He noted that the primary challenge lies in distinguishing genuine users from automated bots, particularly as real citizens inadvertently amplify manufactured messaging. While the data does not suggest that the bot campaigns altered the ultimate election results, they expose an escalating threat industry designed to erode public confidence in democratic processes.
