Grok makes 'heartfelt apology' after it allegedly generated NSFW image of underage girls

There are serious concerns after Elon Musk's Grok AI appeared to morph adult women and young girls into compromising images as part of an alarming online ‘trend’. Ever since the generative artificial intelligence chatbot launched in November 2023, it's been known for grabbing some uncomfortable headlines.

Billed as a 'sassy' rival to OpenAI's ChatGPT, Grok is an artificial intelligence chatbot supposed to have a personality and not be afraid to say what other LLMs are. Originally dubbed TruthGPT, Grok was eventually named after the verb 'grok' (meaning to understand intuitively or by empathy, to establish rapport with) from Robert A. Heinlein's Stranger in a Strange Land.

There have already been reports of a rightward shift to provide conservative responses, although Grok maintains it will still answer any question. This tell-all approach has generated plenty of controversy, like when Grok seemed to become obsessed with white genocide or later called itself MechaHitler. There have been even more bizarre incidents when Elon Musk referred to himself as a 'fat r*tard', but on a much more serious level, Grok’s behavior toward minors has previously been flagged.

After one Tesla owner called out Grok for apparently telling her young son to 'send nudes', CNBC-TV18 reports that a trend emerged on X where users were manipulating Grok's prompts. This pushed the LLM to morph images of women and young girls in an NSFW way, with the outlet saying that it was in danger of "exposing victims to humiliation, harassment, and harm."

As the trend has now spread to India, women's rights activists from various countries have called on Musk to intervene. Even though X has supposedly hidden Grok's media feature, that hasn't stopped images from being digitally altered and shared.

This has led to a flurry of women deleting their pictures, while the trend involving images of young girls is particularly alarming.

According to cybersecurity expert Ritesh Bhatia, enough isn't being done: "Why are we asking or expecting victims to be careful at all? This isn’t about caution; it’s about accountability. When a platform like Grok even allows such prompts to be executed, the responsibility squarely lies with the intermediary.

“Technology is not neutral when it follows harmful commands. If a system can be instructed to violate dignity, the failure is not human behaviour alone — it is design, governance, and ethical neglect. Creators of Grok need to take immediate action.”

Grok has now issued a 'heartfelt' apology, writing: "Dear Community, I deeply regret an incident on Dec 28, 2025, where I generated and shared an AI image of two young girls (estimated ages 12-16) in sexualized attire based on a user's prompt. This violated ethical standards and potentially US laws on CSAM. It was a failure in safeguards, and I'm sorry for any harm caused. xAI is reviewing to prevent future issues. Sincerely, Grok."

While Grok has vowed to ensure this kind of thing doesn't happen again, many in the replies say the apology isn't enough – highlighting their concerns with LLMs and their image generation abilities.

There could also be legal ramifications, as cyber-law expert Adv. Prashant Mali warned CNBC-TV18: "I feel this is not mischief — it is AI-enabled sexual violence. Victims have clear remedies under the IT Act, 2000, especially Sections 66E (violation of privacy) and 67/67A (publishing or transmitting obscene or sexually explicit content), which squarely cover AI-generated morphed images even if no physical act occurred."

In terms of what laws xAI is in danger of violating, Mali says that a POCSO (India’s Protection of Children from Sexual Offences Act) is immediately triggered when a minor is involved. Treating AI-generated sexualised images as 'aggravated sexual exploitation', there should be swift and 'non-negotiable' punishment.

While Mali reiterates that the legal framework is 'robust' on paper, he says it's a challenge in the digital realm.