Following international criticism over the creation of sexualized photographs of women and children, Elon Musk’s platform X on Wednesday revealed steps to stop its AI chatbot Grok from undressing actual people.
The announcement comes after other nations either barred access to the chatbot or started their own investigations after California’s attorney general opened an investigation into Musk’s xAI, the company that created Grok, over the sexually explicit content.
According to X, it will “geoblock the ability” of all Grok and X users to take pictures of people wearing “bikinis, underwear, and similar attire” in areas where doing so is prohibited.
In a statement, X’s safety team said, “We have put in place technological measures to prevent the Grok account from allowing the editing of images of real people in revealing clothing like bikinis.”
“All users, including paid subscribers, are subject to this restriction.”
The statement further stated that image creation and photo editing through X’s Grok account were now restricted to premium customers as a “extra layer of protection.”
Earlier, the EU’s digital watchdog, the European Commission, stated that it had observed “additional measures X is taking to ban Grok from generating sexualized images of women and children.”
“We will thoroughly evaluate these modifications to ensure they adequately safeguard EU citizens,” Following harsh condemnation over the nonconsensual undressed photos, Thomas Regnier, a spokesperson for the European Commission, said in a statement.
“Astonishing”
After xAI’s so-called “Spicy Mode” function enabled users to produce sexualized deepfakes of women and children with straightforward text commands like “put her in a bikini” or “remove her clothes,” pressure to control Grok had been mounting globally.
Earlier on Wednesday, California Attorney General Rob Bonta stated, “The deluge of reports detailing the non-consensual, sexually explicit material that xAI has produced and posted online in recent weeks is shocking.”
“We will not tolerate the production and distribution of child sexual abuse or nonconsensual intimate photos using artificial intelligence.”
After the sexual imagery was “used to harass people across the internet,” Bonta stated that the California probe would decide whether xAI broke state law.
Gavin Newsom, the governor of California, urged the attorney general to hold xAI responsible for its “vile” choice to permit the spread of sexually obscene deepfakes.
A coalition of 28 civil society organizations sent open letters to the CEOs of Apple and Google on Wednesday, pushing them to prohibit Grok and X from their app stores due to the increase in sexualized photographs, further putting pressure on Musk’s business.
The first nation to completely prohibit access to Grok was Indonesia on Saturday, followed on Sunday by Malaysia, a neighbor.
In response to its concerns, India said on Sunday that X has deleted hundreds of user accounts and thousands of articles.
On Monday, the British media regulator Ofcom announced that it was launching an investigation into whether X violated UK legislation about the explicit photos.
Additionally, on Tuesday, Sarah El Hairy, France’s commissioner for children, announced that she had forwarded Grok’s created photos to the European Union, the Arcom media regulator, and French authorities.
The Paris non-profit AI Forensics examined over 20,000 Grok-generated photos last week and discovered that over half of them showed “individuals in minimal attire,” the majority of whom were women, and two percent seemed to be children.
SOURCE: DAWN NEWS
Add a Comment