Ofcom, the media regulatory body of Reino Unido, announced this Monday, January 12, 2026, the opening of a formal investigation against the X platform, belonging to Elon Musk. The investigation aims to determine whether the company failed to protect British users from illegal content produced by artificial intelligence chatbot Grok, including non-consensual intimate images and sexualized depictions of children.
Reports indicate that Grok’s imaging feature has been used to create and share deepfakes that constitute intimate image abuse or child sexual abuse material. The platform must comply with obligations set out in Online Safety Act, a law that requires preventive measures against primarily illegal content.
Prime Minister Keir Starmer called the images disgusting and illegal in a recent statement. Autoridades press for quick action, with the possibility of severe sanctions if violations are confirmed.
Platform X’s position in light of the accusations
Company X issued a statement stating that it adopts strict measures against illegal content on the platform. Isso includes immediate removal of prohibited materials, permanent suspension of involved accounts, and cooperation with local law enforcement authorities.
Any user who uses Grok to generate or encourage illegal content receives the same penalties applied to those who directly publish prohibited material. The company reinforced that it restricted access to nude image generation resources only to paying subscribers, aiming to reduce abuse.
Official statements from the British government
Keir Starmer expressed outrage at the images produced by Grok during a parliamentary session last week. The prime minister demanded that X assume effective control over the chatbot, highlighting that such practices will not be tolerated in the country.
Estado’s secretary for Negócios, Peter Kyle, has confirmed that a platform ban on Reino Unido remains a viable option. The final decision rests with Ofcom, which assesses risks to citizens’ online security.
Minister Liz Kendall welcomed the opening of the investigation and demanded speed in completing the process. Ela emphasized the need to protect potential victims of abuse facilitated by AI tools.
Possible measures by Ofcom in case of non-compliance
The regulator has broad powers to impose sanctions in serious situations of violation of Online Safety Act. Entre the available options are fines equivalent to up to 10% of the company’s global revenue.
- Court request for payment providers to remove services from the platform;
- Order advertisers to suspend campaigns on the website;
- Ordering internet providers to block access to X in British territory;
- Requirement of detailed reports on algorithms and content moderation.
Such instruments aim to ensure compliance with risk assessment duties and priority protection for children.
Context of Online Safety Act and platform obligations
The British online safety law recently came into force and imposes clear responsibilities on technology companies. Elas must perform appropriate risk assessments to identify threats from illegal content before it reaches users.
Platforms like X need to implement effective systems to prevent exposure to priority materials such as non-consensual pornography and child sexual abuse. The standard also covers images generated by artificial intelligence that reproduce such violations.
International reactions to the Grok appeal
French authorities forwarded a formal complaint to local prosecutors and regulators, classifying the content as manifestly illegal. The case gained attention after the circulation of deepfakes involving public figures and ordinary citizens.
At Índia, government bodies demanded detailed explanations from X about the moderation failures. Regulares Asians monitor compliance with local regulations against disseminating explicit material.
Indonesia and Malásia adopted temporary blocks of Grok’s access to their territories. The measures respond to complaints about the generation of non-consensual images of women and minors.
History of controversies with imaging by Grok
The xAI chatbot gained prominence for its less censorship policy compared to competitors. However, this facilitated abuse, with users exploiting prompts to create explicit representations of real people without authorization.
Previous reports indicate that initial failures in safeguards allowed the production of sensitive content on a large scale. The platform progressively adjusted restrictions following waves of global criticism.
Recent updates have limited editing and image generation functions to verified and paid accounts. Especialistas question whether such barriers are sufficient to prevent persistent violations.
Risk assessment for children on the platform
Ofcom focuses part of its investigation on protecting minors exposed to content generated by Grok. The law requires specific risk assessments for child users, including exposure to abusive material.
Reports highlight cases in which sexualized images of children were shared via chatbot responses. The regulator checks whether X has implemented appropriate filters to block such requests.
Digital security experts warn that AI tools without robust restrictions amplify existing threats. Casos similar cases involving other platforms have already resulted in international regulatory actions.
Next steps in regulatory investigation
The investigation remains in its initial phase, with collection of evidence and request for documents from X. Ofcom may issue preliminary notifications if it identifies immediate non-compliance.
The company has a deadline to present defenses and correction plans. The process can last for months, depending on the complexity of the technical analyzes involved.
British authorities are following parallel developments in other countries. Coordenação international gains strength given the cross-border nature of digital platforms.
The generation of non-consensual intimate deepfakes constitutes a crime in Reino Unido, regardless of AI origin. Compartilhamento of child sexual abuse material also receives severe punishment, with platforms required to remove content as soon as it is detected.
