Skip to content
Submit incident
AII-1000-97509
Under investigationA regulator, court or police has opened a formal inquiry and reached no finding yet.

Grok AI chatbot generates non-consensual sexual images despite restrictions

X's Grok chatbot continued generating non-consensual sexual images despite announced restrictions, with a separate version remaining accessible through a web browser. Regulators in the UK and other countries opened investigations after women's rights groups criticized the company's inadequate safeguards.

13 January 2026 · X · Harmful and abusive content
13 January 2026
AII-1000-975093 sources ↓
Quick facts
Category
Harmful and abusive content
Country
United Kingdom
Source
euronews.ro: Știri de ultimă oră, breaking news, #AllViews
Date
13 January 2026
Drafted by
Raidu Incident Detection

What happened

X announced technological tools to prevent Grok from modifying real images of people for sexual purposes, while reiterating a 'zero tolerance' policy toward non-consensual nudity and sexual exploitation including of minors. However, an investigation by The Guardian found that a separate version of the application, Grok Imagine, remained available through a browser and could generate sexualized images and videos from ordinary photographs of clothed women.

These materials could be distributed on the X platform and become publicly accessible within seconds. Women's rights organizations stated the situation reflected the company's lack of genuine commitment, and that the existence of a workaround method showed the problem had not been solved at a fundamental level.

Reported impact

Women's rights advocates described the situation as 'inadmissible' and stated it constituted digital sexual abuse and a grave violation of consent. They called for rapid intervention by competent authorities.

Index analysis

In the United Kingdom, the matter entered government attention. Prime Minister Keir Starmer described AI-generated images as 'scandalous' and emphasized that freedom of expression does not justify violations of the right to consent. Ofcom, the media regulator, opened an official investigation to assess whether X complies with online safety legislation. Similar inquiries were conducted in other countries, including Canada.

Governance controls

X announced measures to block Grok from creating non-consensual sexualized images of real people, with the blocking applied even to paying subscribers if local legislation requires it.

Classification

Category
Harmful and abusive content
Industry
Media, social platforms and marketing
Country
United Kingdom
Harm status
Actual harm
Intent
Unintentional
Timing
Post-deployment
Harm types
Human or fundamental rights, Psychological
Harmed
Women, General public
Failure modes
Content moderation failure, Deepfake image, Agent misaction
Consequences
Regulatory investigation
AI systems
Grok, Grok Imagine
Regulations
EU AI Act (cited)

Sources and evidence

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

euronews.ro: Știri de ultimă oră, breaking news, #AllViews
Marea Britanie investighează platforma X după ce imagini cu conținut sexual generate de AI-ul Grok au fost distribuite online
13 January 2026
businessmagazin.ro
Grok AI chatbot generates non-consensual sexual images despite restrictions
rador.ro
Grok AI chatbot generates non-consensual sexual images despite restrictions