Skip to content
Submit incident
AII-1000-69513
ReportedReported by fewer than three independent publishers, with no official finding.

X's algorithm recommends explicit content to UK teenagers

A study by the Center for Countering Digital Hate found that X's recommendation algorithm and content moderation systems exposed teenage users in the UK to sexually explicit material, weak community safeguards, and unsolicited contact from adults, reportedly breaching the Online Safety Act.

14 April 2026 Β· X Β· Harmful and abusive content
14 April 2026
AII-1000-695132 sources ↓
Quick facts
Category
Harmful and abusive content
Country
United Kingdom
Source
The Independent
Date
14 April 2026
Drafted by
Raidu Incident Detection

What happened

The Center for Countering Digital Hate created two test accounts posing as 13-year-old UK users and tested X's content moderation across search, recommendations, communities, and direct messaging.

When researchers searched for sexual terms such as 'sex, porn, boobs, and hentai', 80 per cent of searches returned explicit results within the first ten results containing media, with no age verification, content warning, or filtering step.

After explicit searches, the For You algorithm altered to recommend graphic images, with 30.5 per cent of recommended posts being explicit. Teen accounts joined 15 out of 20 adult sexual communities with ease, and received direct message requests from adult accounts, including unsolicited videos of a man masturbating.

Reported impact

The report warned of potential breaches of the Online Safety Act, with X's weak safeguards exposing teenagers to explicit content and possible direct sexual contact from adults.

Index analysis

The failure appears systematic: search moderation did not filter explicit results; recommendation algorithms amplified sexual content after initial searches; community access controls were bypassed; and default direct messaging protections were circumvented.

Classification

Category
Harmful and abusive content
Industry
Media, social platforms and marketing
Country
United Kingdom
Harm status
Potential harm
Intent
Unknown
Timing
Post-deployment
Harm types
Psychological, Human or fundamental rights
Harmed
Children
Failure modes
Content moderation failure, Unsafe recommendation
AI systems
X's recommendation algorithm

Sources and evidence

This record was researched and written by the Index, based on the following original reporting.

The Independent
Explicit content consistently recommended to 13-year-olds on X, report finds
15 April 2026
Hindustan Times
Study finds Elon Musk's X exposing young teens to pornographic content: Report
15 April 2026
Oct 2026Mexico's Women's Ministry Condemns Grok for Generating Non-Consensual Sexual ImagesConfirmedHarmful and abusive contentHora Cero WebEl UniversalAug 2026Harry and Meghan criticize Musk's opposition to Minnesota AI deepfake lawReportedHarmful and abusive contentUSA TodayAOL.comMay 2026Spanish Ombudsman Investigates Harmful AI-Driven Social Media Content Affecting MinorsUnder investigationHarmful and abusive contentCadena SEREL PAÍSCanarias7+4May 2026EU investigation into X's Grok AI for generating sexual content involving minorsUnder investigationHarmful and abusive contentmfor.hu - Menedzsment FórumPrivÑtbankÑr.huMar 2026Dutch Court Bans Grok from Generating Non-Consensual Sexual Deepfakes; EU Parliament Votes BanConfirmedHarmful and abusive contentagenceurope.euCNEWSDH.be+3