4 months ago
1
Elon Musk's level X connected Wednesday (January 14, 2026) announced measures to forestall its AI chatbot Grok from undressing images of existent people, pursuing planetary backlash implicit its procreation of sexualized photos of women and children.
The announcement comes aft California's lawyer wide launched an probe into Mr. Musk's xAI — the developer of Grok — implicit the sexually explicit worldly and aggregate countries either blocked entree to the chatbot oregon launched their ain probes.
X said it volition "geoblock the ability" of each Grok and X users to make images of radical successful "bikinis, underwear, and akin attire" successful those jurisdictions wherever specified actions are deemed illegal.
"We person implemented technological measures to forestall the Grok relationship from allowing the editing of images of existent radical successful revealing covering specified arsenic bikinis," X's information squad said successful a statement.
"This regularisation applies to each users, including paid subscribers."
In an "extra furniture of protection," representation instauration and the quality to edit photos via X's Grok relationship was present lone disposable to paid subscribers, the connection added.
The European Commission, which acts arsenic the EU's integer watchdog, earlier said it had taken enactment of "additional measures X is taking to prohibition Grok from generating sexualised images of women and children."
"We volition cautiously measure these changes to marque definite they efficaciously support citizens successful the EU," European Commission spokesperson Thomas Regnier said successful a statement, which followed crisp disapproval implicit the nonconsensual undressed images.
‘Shocking’
Global unit had been gathering connected xAI to rein successful Grok aft its alleged "Spicy Mode" diagnostic allowed users to make sexualized deepfakes of women and children utilizing elemental substance prompts specified arsenic "put her successful a bikini" oregon "remove her clothes."
"The avalanche of reports detailing the non-consensual, sexually explicit worldly that xAI has produced and posted online successful caller weeks is shocking," California Attorney General Rob Bonta said earlier connected Wednesday (January 14).
"We person zero tolerance for the AI-based instauration and dissemination of nonconsensual intimate images oregon of kid intersexual maltreatment material."
Mr. Bonta said the California probe would find whether xAI violated authorities instrumentality aft the explicit imagery was "used to harass radical crossed the internet."
California Governor Gavin Newsom said that xAI's "vile" determination to let sexually explicit deepfakes to proliferate prompted him to impulse the lawyer wide to clasp the institution accountable.
Further adding unit onto Mr. Musk's institution connected Wednesday (January 14), a conjugation of 28 civilian nine groups submitted unfastened letters to the CEOs of Apple and Google, urging them to prohibition Grok and X from their app stores amid the surge successful sexualized images.
Indonesia connected Saturday (January 10) became the archetypal state to artifact entree to Grok entirely, with neighbouring Malaysia pursuing connected Sunday (January 11).
India said connected Sunday (January 11) that X had removed thousands of posts and hundreds of idiosyncratic accounts successful effect to its complaints.
Britain's Ofcom media regulator said connected Monday (January 12) it was opening a probe into whether X failed to comply with U.K. instrumentality implicit the intersexual images.
And France's commissioner for children Sarah El Hairy said connected Tuesday (January 13) she had referred Grok's generated images to French prosecutors, the Arcom media regulator and the European Union.
Last week, an investigation of much than 20,000 Grok-generated images by Paris non-profit AI Forensics recovered that much than fractional depicted "individuals successful minimal attire" — astir of them women, and 2 percent appearing to beryllium minors.
