Illustrative photo. Source: Dreamstime.com
The artificial intelligence model Grok, developed by Elon Musk, received technological restrictions that prevent the editing of photos of real people in indecent poses. This measure, applied with priority in regions where such manipulations are illegal, comes as a direct response to the global outrage caused by deepfake images with sexual undertones, according to the BBC.
“We have implemented technological measures to prevent the Grok account from allowing the editing of images of real people in provocative outfits. This restriction applies to all users, including paying subscribers,” said an announcement on X, which operates the Grok AI tool.
The decision was made public shortly after California's attorney general launched a formal investigation into the use of AI to create sexual deepfakes, including illegal content with minors.
“We are now geo-blocking the ability of all users to generate images of real people in bikinis, lingerie and similar clothing through their Grok account and within Grok in X in jurisdictions where this is illegal,” X said in a statement on Wednesday.
The platform reaffirmed that image editing through Grok remains a privilege reserved exclusively for subscribers. This financial conditioning works as a safety filter, with X arguing that removing anonymity will allow the prosecution of any user who uses AI to break the law.
Elon Musk clarified, on Wednesday, the limits of the new algorithm. He stated that by enabling NSFW (Not Safe For Work) settings, the Grok chatbot will only be able to generate partial nudity and exclusively for fictional characters. Musk emphasized that Grok will be limited to what is typically allowed in an R-rated movie (not allowed to minors without parental consent).
“This is the de facto standard in America. This will vary in other regions depending on the laws in each country,” the billionaire told X.
Prior to the implementation of the new restrictions, Musk took a defiant stance, defending the platform's right to generate content without censorship. He accused Grok's critics of only using the online safety argument as a pretext to “suppress free speech.”
The wave of criticism of Grok's editing function intensified globally, culminating in drastic decisions in Southeast Asia. Malaysia and Indonesia marked a first over the weekend, becoming the first countries to block access to Platform X's AI tool. The decision came after numerous users complained about the generation of explicit and non-consensual content by manipulating real photos.
Photo source: Dreamstime.com
