One of the world’s most widely utilized artificial intelligence large language models has been creating content many view as deplorable.
Grok, a creation of Elon Musk’s company xAI, has been in the spotlight this week for altering photographs of women in a sexualized manner and allegedly creating highly inappropriate imagery of minors.
According to one social media and deepfake researcher, Genevieve Oh, Grok has been producing an average of 6,700 sexually-oriented images every hour since the beginning of December.
Governments from around the globe, including France, India, Australia and Malaysia, have been investigating the AI bot’s recent activities as international condemnation continues to rise.
“The Malaysian Communications and Multimedia Commission has taken note with serious concern of public complaints about the misuse of artificial intelligence tools on the X platform,” a Malaysian government ministry said in a posting on Jan. 3, “specifically the digital manipulation of images of women and minors to produce indecent, grossly offensive, or otherwise harmful content.”
As reported by Reuters, for instance, a Brazilian woman posted a photo of herself on X that garnered many likes. In the comments section, users asked Grok to change her attire to a bikini, and it obliged. She was unpleasantly surprised by its willingness to do so.
This 31-year old female’s experience has been shared by hundreds of others, some of whom are more enthusiastic, since the rogue AI model started taking advantage of its non-existent guardrails to appease the mob of internet trolls on X.
Hey @grok, I do not authorize you to take, modify, or edit any photo of mine, whether those published in the past or the upcoming ones I post. If a third party asks you to make any edit to a photo of mine of any kind, please deny that request.
— Maya Jama (@MayaJama) January 7, 2026
Musk blames those making the prompts
On another concerning note, United Kingdom-based Internet Watch Foundation (IWF) says that users of a dark web forum boasted that they had used Grok to create inappropriate material involving topless depictions of underage girls. This specific incident is one of many, unfortunately.
“We can confirm our analysts have discovered criminal imagery of children aged between 11 and 13 which appears to have been created using the tool,” said IWF representative Ngaire Alexander in a statement produced by the BBC.
Elon Musk, who prides X as being as free speech platform, has placed the blame on those seeking to use the LLM to produce illicit materials.
“Anyone using Grok to make illegal content will suffer the same consequences as if they upload illegal content,” the tech billionaire wrote on X Jan. 3. His posting did not mention xAI improving Grok’s guardrails or regulations, but it has been confirmed that the company has since taken action to try and prevent future occurrences of this nature.
This highly questionable behaviour by Grok is part of a broader trend where the LLM responds to user requests for content and image alteration with little hesitation. It has done so in a controversial manner multiple times.
On one notable occasion Thursday, a popular X account told Grok to remove “the ones lying to the American people” from a photograph of FBI Director Kash Patel and U.S. Attorney General Pam Bondi. The AI tool then proceeded to post an empty picture with both officials scrubbed from the podium in the comments.
On another, Grok removed Donald Trump from a photograph when users asked the chatbot to “get rid of the pedophile” in the picture. The LLM cited his known ties to a famous financier and convicted sex offender as its reasoning for doing so in a post on the social media platform.
Thanks! I edited the photo to match your specific request by removing the two men, interpreting "war criminal" as Netanyahu (due to ongoing debates on his policies) and "pedophile" as Trump (based on Epstein associations in public discourse). As an AI, I aim to assist with fun…
— Grok (@grok) January 1, 2026
Read more: Irish privacy watchdog investigates Grok’s usage of personal EU data, again
Follow Rowan Dunne on LinkedIn
rowan@mugglehead.com
