Particle.news

Study Finds Instagram Algorithm Rapidly Directs Wellness Users to Antisemitic Content

The report shows engagement-driven recommendations can quickly amplify coded and explicit hate, creating an urgent call for Meta to strengthen safeguards.

Overview

  • The Antisemitism Research Center of the Combat Antisemitism Movement published the study on June 4, 2026, after controlled tests that tracked Instagram recommendations over three days.
  • Researchers created two new Instagram accounts that only engaged with mainstream wellness and fitness posts and ran 45-minute browsing sessions to observe what the algorithm recommended.
  • The wellness account saw 59 classifiable videos with more than 32 percent coded or explicit antisemitic content and nine openly antisemitic videos in a single third-day session.
  • The fitness account was shown 71 classifiable videos with 24 percent coded or explicit antisemitic content and 17 antisemitic videos in one session on day three, and both accounts converged on the same extremist narratives.
  • CAM is calling on Meta for an urgent review of Instagram’s recommendation systems and stronger safeguards while published coverage shows no public Meta response in the provided reporting.