AI Grok has found itself at the centre of an international scandal over sexualised dipfakes

The scandal was sparked by the launch of the "edit image" button, which allowed users to edit images based on text prompts, including requests such as "undress her" or "put a bikini on her". This sparked a flurry of complaints from users, human rights activists and authorities in various countries writes Techxplore.

EU and UK reaction

The European Commission said it is scrutinising complaints about Grok's work. Thomas Renier, the European Commission's digital affairs spokesperson, called the situation unacceptable:

"Grok offers a so-called 'spicy mode' that generates sexually explicit content, including images with childish features. This is not 'spicy.' It's illegal. And it has no place in Europe."

In the UK, media regulator Ofcom said it was urgently calling on X and xAI to explain what measures have been taken to protect users. The agency has not ruled out launching a formal investigation.

Investigations and pressure from countries

France, India and Malaysia have also reacted to the developments. In France, prosecutors expanded the investigation into X to include new charges related to the possible distribution of child pornography via AI.

Indian authorities demanded that X remove the illegal content, block the offenders and report back within 72 hours. The deadline has passed, but there is no official information about the company's response yet.

Malaysia's communications regulator said it was "gravely concerned" and signalled its intention to summon X representatives for an explanation.

"It's frightening and illegal."

X users have also been vocal in their opposition to the tool. Malaysian lawyer Azira Aziz said Grok altered her photo without consent by changing her into a bikini.

"Harmless experimentation with AI is one thing. But when the technology is used as a form of gender-based violence against women and children - it must be firmly stopped," she told AFP in a comment.

A statement by Ashley St Clair, the mother of one of Ilon Musk's children, who wrote that Grok was "undressing her baby pictures", also caused resonance.

Grok and xAI's response

Amid the scandal, Grok acknowledged problems with its security mechanisms and said it was urgently fixing the vulnerabilities. The company emphasised that child sexual abuse material is strictly prohibited.

Grok had previously apologised for generating an image of two underage girls in sexualised clothing at the request of a user.

However, xAI's official response to AFP's request was succinct and contained only the phrase: "Legacy Media Lies".

This is not the first scandal around Grok

This is not the first time criticism around Grok has been mounting. Previously, the AI was accused of spreading misinformation about the war in Gaza, the conflict between India and Pakistan, and the shooting in Australia.

Experts note that the situation intensifies the debate on the need for stricter regulation of generative AI and the responsibility of platforms for the content created by their tools.