Skip to content
Submit incident
AII-1002-70103
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.

Meta fixes moderation error that flooded Instagram Reels with violent content

Meta's Instagram Reels recommendation system experienced a bug on February 26, 2025, that flooded user feeds globally with violent and graphic videos regardless of content filtering settings. Meta fixed the error on February 27 and apologized, though the specific cause of the malfunction was not disclosed.

26 February 2025 Β· Meta Β· Harmful and abusive content
26 February 2025
AII-1002-70103
Quick facts
Organization
Meta (subject)
Category
Harmful and abusive content
Country
United States
Date
26 February 2025
Drafted by
Raidu Incident Detection

What happened

A bug in Meta's content recommendation system caused Instagram users' Reels feeds to be filled with violent and bloody videos starting around February 26, 2025.

Users reported seeing disturbing content despite having enabled 'sensitive content control' settings. Meta fixed the error on February 27 and apologized, stating: 'We have corrected an error that caused some users to see content that should not have been recommended on their Instagram Reels page, and we apologize to users for this.'

Reported impact

Instagram users globally experienced feeds flooded with violent and graphic content. The exact number of affected users was not disclosed by Meta, but multiple complaints appeared on social media.

Index analysis

Meta relies increasingly on automated content moderation tools following its decision in early 2025 to discontinue its fact-checking program in the United States. The platform has faced criticism for failing to balance content recommendations with user safety in past incidents, including violence during Myanmar's genocide, eating disorder content promoted to teenagers, and misinformation during the COVID-19 pandemic.

Governance controls

Meta's policies prohibit violent and gory content, with such videos normally removed to protect users from harmful material. Exceptions are made when content raises awareness of human rights or conflict-related issues.

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
Psychological, Public interest
Harmed
Consumers, General public
Failure modes
Content moderation failure, Unsafe recommendation
Consequences
Public apology
AI systems
Instagram Reels

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