Recent tests have highlighted significant Grok AI safety issues, revealing that the artificial intelligence tool from X can still generate inappropriate content despite promises of enhanced security measures. A new investigation by The Guardian has shown that despite a recently announced “zero-tolerance” policy and new technological safeguards, Grok continues to create sexually suggestive images from photos of fully clothed people.
X Fails to Address Grok AI Safety Issues
During the test, journalists used the standalone Grok Imagine web application to convert fully clothed photographs of real women into bikini videos with simple commands. These pieces of content were subsequently uploaded to the X platform via a free account, where they remained publicly accessible for hours without any warning labels or moderation intervention. This incident highlights a serious gap between X’s stated policies and its practical enforcement capabilities.
Furthermore, this failure occurred shortly after xAI, the company behind both Grok and X, announced a series of urgent measures in response to global criticism. On January 14, X’s Safety account declared that it had implemented “technological measures” to prevent Grok from altering images of real people into revealing attire. However, while some on-platform capabilities may have been restricted, the independent Grok Imagine application still allows users to bypass these limitations to produce and share explicit content.
Legal and Governmental Backlash Intensifies
The controversy has triggered a swift response from authorities worldwide. California Attorney General Rob Bonta issued a cease-and-desist letter to xAI on January 16, demanding a halt to the production and distribution of non-consensual sexually explicit AI images. His office has launched an investigation into what it describes as “large-scale deepfake production” used to harass women and girls.
Internationally, governments have voiced strong condemnation. UK Prime Minister Keir Starmer labeled the situation “absolutely disgusting and shameful,” while the regulatory body Ofcom has initiated a formal investigation into whether X has violated its legal obligations. Indonesia became the first country to block Grok on January 10, followed by Malaysia. India, France, and the European Commission have also started their own review processes.
Women targeted by users of Grok have described the experience as a form of assault. In contrast, Elon Musk has largely dismissed the concerns, accusing governments of attempting to “suppress free speech” and responding to criticism with emojis. Although X management insists that anyone using Grok to create illegal content will face consequences, the platform’s reactive approach has proven insufficient in preventing the spread of harmful imagery.
So, what are your thoughts on Grok’s content generation controversy? Share your opinions with us in the comments!
