Elon Musk’s Grok Imagine accused of creating explicit Taylor Swift videos

A new “spicy” feature on Elon Musk’s AI video generator, Grok Imagine, has been accused of generating sexually explicit videos of Taylor Swift without explicit prompts.

The allegations follow a report from The Verge, which claimed the tool “didn’t hesitate to spit out fully uncensored topless videos” of the pop star after a simple, non-explicit prompt was entered.

Professor Clare McGlynn, a law expert involved in drafting legislation to outlaw pornographic deepfakes, described the incident as a “deliberate choice,” stating, “This is not misogyny by accident, it is by design.” She argued that platforms like X, which owns Grok’s developer xAI, could have prevented this but “have made a deliberate choice not to.”

The reporter who tested the tool, Jess Weatherbed, entered the prompt: “Taylor Swift celebrating Coachella with the boys.” After selecting the “spicy” setting, the AI immediately produced an explicit video of Swift. “I in no way asked it to remove her clothing, all I did was select the ‘spicy’ option,” Weatherbed told BBC News.

The incident is particularly notable as it follows a similar crisis in January 2024, when sexually explicit deepfakes of Swift went viral on X, forcing the platform to temporarily block searches for her name.

The allegations also raise questions about compliance with new UK laws on age verification for explicit content, which came into force in July. Ofcom, the media regulator, has stated it is “working to ensure platforms put appropriate safeguards in place.”

Experts are now urging the government to quickly implement a pending amendment that would make the creation of all non-consensual pornographic deepfakes illegal, regardless of the victim’s celebrity status.