Overview
- FIFA said on Wednesday that its Social Media Protection Service scanned more than six million posts in the group stage, with artificial intelligence flagging about 225,000 items for human review and moderators verifying 89,000 as abusive.
- Racial abuse accounted for 11 percent of detected offensive messages, a rise from the comparable stage in 2022, and automated tools hid roughly 181,000 hateful comments from team accounts.
- About 1,000 user accounts were escalated for deeper investigation and FIFA identified over 100 examples that meet legal thresholds for preparing case files to share with law enforcement.
- FIFA and its data partner use a two-step system of AI screening followed by human analysts to filter spam, bot content and evolving tactics such as weaponized emojis and AI-generated images.
- The jump in volume is tied to the expanded 48-team tournament and more matches, which increases public posts to monitor and raises challenges for attribution and cross-jurisdictional legal action that affect players, coaches and officials.