Select Language:
On Wednesday, Elon Musk’s platform X announced new measures to prevent its AI chatbot Grok from creating undressing images of real people, in response to worldwide criticism over sexually explicit photos of women and children. This move follows California’s attorney general opening an investigation into Musk’s xAI, the creator of Grok, over such explicit content, with several countries either blocking access or conducting their own probes.
X stated it will “geoblock the ability” for all users within certain regions to generate images of individuals in bikinis, underwear, or similar clothing where such actions are illegal. The safety team explained that they’ve implemented technological safeguards to stop Grok accounts from editing images of real people in revealing outfits, and these restrictions apply even to paid subscribers.
Additionally, image creation and editing features for Grok are now exclusive to paid subscribers, adding an extra layer of security. The European Commission, serving as the EU’s digital watchdog, acknowledged the steps X is taking to curb the generation of sexually explicit images of women and children, promising to evaluate whether these measures adequately protect European citizens.
Pressure mounted on xAI after its “Spicy Mode” feature allowed users to produce sexualized deepfakes of women and minors with simple prompts like “put her in a bikini” or “remove her clothes.” California Attorney General Rob Bonta condemned the recent reports of non-consensual, sexually explicit content produced and shared online, emphasizing that there is zero tolerance for AI-facilitated creation and distribution of such material, including child sexual abuse imagery. The investigation will determine if xAI violated state laws by facilitating harassment through these images.
California Governor Gavin Newsom criticized xAI’s decision to enable the proliferation of sexually explicit deepfakes, urging accountability. Meanwhile, a coalition of 28 civil society groups sent open letters to Apple and Google CEOs, urging them to remove Grok and X from their app stores amid the surge in sexualized content.
Indonesia became the first country to fully block access to Grok, doing so last Saturday, followed by Malaysia on Sunday. India reported removing thousands of posts and hundreds of user accounts in response to complaints. The UK’s Ofcom announced on Monday it would investigate whether X violated local laws regarding sexually explicit imagery.
A recent analysis by the nonprofit AI Forensics, based in Paris, examined over 20,000 images generated by Grok. Findings revealed that more than half depicted individuals in minimal clothing—most being women and approximately 2% appearing to be minors.
ChatGPT
Perplexity
Gemini AI
Grok AI
