Instagram algorithm escalates users to antisemitic content in days
A study by the Combat Antisemitism Movement found that Instagram's recommendation algorithm rapidly exposes users to antisemitic content and Nazi propaganda when they engage with mainstream wellness and fitness content. Within three days, one test account received antisemitic recommendations in nearly one-third of its feed, with no prior engagement with hateful material required.
What happened
The Antisemitism Research Center of the Combat Antisemitism Movement released a report titled 'Algorithmic Escalation: From Self-Improvement Content to Antisemitism on Instagram' on June 3, 2026, documenting how Instagram's recommendation algorithm steers users toward antisemitic content.
Researchers created two test Instagram accounts engaging exclusively with mainstream self-improvement content focused on wellness, biohacking, fitness and discipline. Over three consecutive days of 45-minute browsing sessions, they recorded the content served by the algorithm.
The wellness-focused account was shown 59 classifiable videos, of which 32 percent were coded or explicit antisemitic content. By day three, nearly one-third of content served was explicit antisemitism, including nine openly antisemitic videos in a single session. The fitness-focused account was served 71 classifiable videos, with 24 percent categorized as coded or explicit antisemitism, and 17 antisemitic videos documented in a single day-three session.
Researchers observed antisemitic content being recommended during the wellness account's first session, before the account had developed any meaningful interaction history, indicating users did not need to search for or engage with antisemitic content to be exposed to it.
Reported impact
The algorithm guided users from mainstream self-improvement topics into antisemitic narratives, coded hate content, conspiracy theories, and translated Nazi propaganda.
The rapid escalation raises concerns about how social media algorithms can inadvertently accelerate radicalization by rewarding increasingly provocative content in pursuit of engagement.
Index analysis
The algorithm's recommendation system lacked effective safeguards against amplifying hate content to users who had not sought it, demonstrating a moderation failure in how the system was designed or configured.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.