Skip to content
Submit incident
AII-1001-82091
ConfirmedA court, a regulator or the organization itself has stated the core facts, or the record cites a court or regulator source.Supporting sources are pending verification.

Instagram Algorithm Malfunction Floods Reels Feeds with Violent and Graphic Content

Instagram's recommendation algorithm malfunctioned in February 2025, flooding users' Reels feeds with violent and graphic content including footage of death and severe injury. Meta apologized and stated the error had been fixed, denying it was related to recent changes to its content moderation policies.

28 February 2025 Β· Meta Β· Harmful and abusive content
28 February 2025
AII-1001-82091
Quick facts
Organization
Meta (subject)
Category
Harmful and abusive content
Country
United States
Date
28 February 2025
Drafted by
Raidu Incident Detection

What happened

Instagram users reported being flooded with violent and graphic content on their Reels feeds in February 2025, including footage of death, gore, animal abuse, and severe injury.

Meta stated the issue resulted from an algorithm malfunction that recommended content that should not have been shown, and that the company fixed the error. Meta said the malfunction was unrelated to recent changes to its content moderation policies.

Reported impact

Users described being exposed to extreme graphic material, with one reporting seeing at least 10 deaths in their Reels feed and another describing the feed as dominated by violent content for 24 hours.

Governance controls

Meta's content guidelines state the company prohibits violent and graphic content. Some users posted screenshots showing 'sensitive content' screens designed to shield users from graphic material, but these protections did not prevent the content from appearing in feeds.

Classification

Organization
Meta (subject)
Category
Harmful and abusive content
Industry
Media, social platforms and marketing
Country
United States
Harm status
Actual harm
Intent
Unintentional
Timing
Post-deployment
Harm types
Human or fundamental rights, Psychological
Harmed
Consumers, General public
Failure modes
Hallucination, Unsafe recommendation
Consequences
Public apology
AI systems
Instagram Reels Algorithm

Sources and evidence

Sources not yet published

This record was researched and written by the Index from original reporting. Its sources are still being verified by an editor, and they are listed here once that check is complete. Until then the record carries no reviewer.

How sources are verifiedSuggest a source for this incident