Elon Musk’s X restricts Grok AI image editing to paid users amid deepfake backlash

Grok

• X restricts Grok's image editing to paid subscribers after deepfake backlash
• Regulators call for action after Grok used to make non-consensual sexual images.
• Non-subscribers can still access Grok's image editing via its app and website

Did our AI summary help?

Elon Musk-owned social media platform X has restricted image editing capabilities of its AI chatbot Grok to paid subscribers following backlash over the creation of sexualised deepfake images. The move comes amid mounting regulatory pressure and criticism that the tool was being misused to generate non-consensual and unlawful content involving real people.

Why X limited Grok’s image editing features

Grok, which users can tag directly on X to respond to prompts, had also allowed image editing based on uploaded photos. Users were able to ask the AI to digitally alter images of people, including requests to remove clothing, without consent. Several individuals affected by such edits described the experience as humiliating and dehumanising, triggering public criticism and concern from online safety experts.

In response, Grok now tells users attempting to edit or generate images on X that these features are limited to paying subscribers. As a result, only users with verified paid accounts, whose names and payment details are on file, can access image editing directly on the platform. Posts reviewed across X suggest that the capability is currently restricted to accounts with the blue tick associated with paid subscriptions.

Regulatory scrutiny intensifies

The decision follows growing pressure from regulators, particularly in the UK. The government has urged media regulator Ofcom to use all available powers against X over concerns related to unlawful AI-generated images. Prime Minister Sir Keir Starmer described the generation of sexualised images of adults and children as disgraceful and unlawful, stating that such content would not be tolerated.

Under the Online Safety Act, Ofcom has the authority to seek court orders that could limit a platform’s ability to operate or raise revenue in the UK. Government sources have indicated that regulators are expected to act decisively if platforms fail to prevent harm linked to AI tools.

How it came to this

Concerns escalated after updates to Grok made image editing easier through text-based prompts on X. The lack of strong safeguards allowed users to exploit the tool to create non-consensual sexualised images using real photographs. Critics argue that insufficient prompt restrictions and detection mechanisms meant harmful content spread before moderation actions were taken, prompting regulatory intervention.

What remains available

Users who do not subscribe to X can still access Grok’s image editing tools through its separate app and website. X has not confirmed whether further technical safeguards will be introduced to prevent abusive uses of the AI.