According to Cyberwell, a nonprofit that works with social platforms to combat antisensitivity online, the trends in memes generated by anti-Semitic AI are spreading rapidly across social media, portraying Jews as greedy, injustice, or delusional.
The “promised three thousand years ago” trend often combines with the Jewish folk song “Hava Nagira” to use AI-generated caricatures of orthodox Jewish men, often combined with Jewish songs, to laugh at Jewish identity and reinforce harmful stereotypes. The memes portray Jewish figures claiming ownership over absurd objects like cities, cars, soda cans, babies, and even the moon, but exaggerate the physical characteristics associated with anti-Semitic ratios.
In particular, since Hamas began its war with Israel on October 7, 2023, various types of anti-Semitic messages have overrun social media networks in recent years. Platforms like Facebook, X, Instagram and Tiktok have been working to improve the moderation of messages that violate standards but frequently violate them. Even Wikipedia, a collaborative online encyclopedia, struggles to combat anti-Semitic bias.
The new looting is meant to be trivial and smirk at the religious and historical ties Jews have on Israel's land, Cyberwell said, forming the core element of Israel's continued conflict with the Palestinians. It means that Jews manipulate ancient claims to assert rule and create phrases for bio, greed and delusion.
One widely shared video posted to Instagram by Mohamed Hadid to 1.5 million followers shows Jews claiming lawn chairs, sandwiches, clocks and babies, and Hasidic Jews “promised 6 million followers 3,000 years ago.
Another shows that a Jewish man, generated by an AI taking a female dog, claims that he was “promised to him three thousand years ago.”
??????? Knn, everything was promised 3000 years ago… pic.twitter.com/zuafu8bqwr
– ❌❌️ (@motulx22) August 7, 2025
Cyberwell says that social media platforms do not currently categorize this generated trend as violating bullying and harassment policies, taking the position that fictional characters fall outside of existing rules. However, the company said the platform policy explicitly prohibits the historically harmful stereotypes and explicitly prohibits robbing the death of certain groups.
“This latest trend highlights a broader challenge for social media platforms. AI tools are making monitoring and controlling large-scale harmful content more difficult.”
“Relying on users to report hatred has proven to be a complete failure, contributing to the wave of real-world violence amplified by anger-prosperous algorithms.
Cohen Montemayor has called for a platform to coordinate policies to better conclude these types of videos.
“If social media platforms don't invest in stronger and faster response mechanisms, as well as efforts to combat revenge porn, this AI's creative abuse will continue to reach billions,” she said. “In the age of AI, slow and reactive trust and safety enforcement and policy updates are no longer acceptable.”
