A recent investigation has uncovered a vast network of anti-Muslim propaganda spread across 30 countries, revealing a large-scale disinformation campaign linked to actors based in Israel. The campaign, which ran from October 2023 to October 2026, generated up to 831 million advertising exposures, with an estimated budget of $8.2 million. Researchers identified the use of artificial intelligence-produced videos, targeted advertisements, and racist content spread on social media platforms.
The study, titled "Manufacturing the 'Great Replacement'," was conducted by Marc Owen Jones, an associate professor at Northwestern University in Qatar, and Sohan Dsouza, a specialist in open-source intelligence. The investigation documented 533 million advertising exposures in 30 countries, with an estimated total of 831 million when including ads not found in archives. The campaign's impact was felt across Europe, North and South America, Asia, and the Middle East.
The researchers found that the campaign used various formats to spread hostile content against Muslims and migrants, including paid social media ads, AI-generated videos, songs, and video game parodies. These messages also fueled anti-Qatar and anti-Iran sentiment. Part of the campaign relied on the "Great Replacement" conspiracy theory, which posits that European populations are being replaced by immigrant populations, particularly Muslims.
The propaganda campaign spread its messages on multiple platforms, including Facebook, YouTube, and TikTok, to reach different audiences. Analysis of reactions to the content showed that it was capable of sparking anti-Muslim sentiment. In a sample of 406 comments posted in 2024 under Facebook ads targeting the UK, 61% supported hostile opinions towards Muslims.
Technical evidence points to actors based in Israel, including an Israeli IP address used for Wikipedia modifications and video metadata with Hebrew file names. The production schedule and time zone changes in digital files also suggest activity based in Israel. However, the researchers note that these indicators do not conclusively identify the campaign's financiers or backers.
The investigation found that the advertising content was sometimes spread during periods corresponding to electoral deadlines in several European countries. In France, the ads contributed to spreading hostile messages about immigration and Islam to a broad audience. This operation highlights the scale of manipulation campaigns on digital platforms, where paid advertising and AI-generated content enable mass dissemination of racist and Islamophobic discourse.
The report sheds light on an international network whose activities reached millions of internet users across three continents over nearly three years. The campaign's use of artificial intelligence and targeted advertising demonstrates the evolving nature of disinformation tactics. Understanding the scope and impact of such operations is crucial in addressing the spread of hate speech and propaganda online.
Key points
- The campaign generated up to 831 million advertising exposures across 30 countries.
- The estimated budget for the campaign was $8.2 million.
- Technical evidence suggests that the campaign was linked to actors based in Israel.
