Elon Musk-owned social media platform X has announced new restrictions on its artificial intelligence model, Grok, barring it from editing images of real people to depict them in revealing clothing in jurisdictions where such content is illegal.

The move follows growing global outrage over the proliferation of sexualised AI-generated deepfakes.

In a statement shared on X, the company said it had introduced technological safeguards to prevent the Grok account from enabling the editing of images of real individuals in revealing clothing.

X explained that the new restrictions were implemented as part of efforts to curb abuse of the artificial intelligence tool.

The announcement came just hours after the Attorney General of California, Rob Bonta, revealed that his office had launched an investigation into the spread of sexualised AI-generated deepfakes, including images involving children, which were linked to Grok’s image-editing capabilities.

“We now geoblock the ability of all users to generate images of real people in bikinis, underwear, and similar attire via the Grok account and in Grok in X in those jurisdictions where it’s illegal,” the company said on Wednesday.

X also restated that only paid subscribers would be permitted to edit images using Grok. The company said this policy would add “an extra layer of protection” by making it easier to identify and hold accountable users who abuse the tool to violate the law or the platform’s policies.

Reacting to the development, the United Kingdom government said it felt “vindicated” following public criticism of X by Prime Minister Sir Keir Starmer, who had earlier challenged the platform to bring the controversial AI tool under tighter control.

With not-safe-for-work settings enabled, Grok remains permitted to generate “upper body nudity of imaginary adult humans (not real ones),” a standard that aligns with R-rated movie classifications in the United States, Musk said in a post on Wednesday. He added that content standards would differ “country by country” in line with local laws.

Musk had previously defended X against mounting criticism, accusing those opposed to the platform’s approach of attempting to “suppress free speech.” He also reposted AI-generated images depicting UK Prime Minister Sir Keir Starmer wearing a bikini, an action that further intensified public backlash.