Just weeks after 54-year-old Elon Musk was called out for making a creepy, juvenile AI video about actor Sydney Sweeney's breasts, he decided to promote the use of her likeness and voice to tout how great his X AI Grok Imagine—a text-to-video feature—is at making deep fakes.
The video, originally posted by another user, featured an AI created Sweeney on a spaceship speaking about Grok videos. The original prompt didn't specify Sweeney by name, leading many to wonder if Musk had altered Grok's responses again.
Musk shared the video with the caption parroting the fake Sweeney's message:
"Grok video is now 10 seconds and the audio is greatly improved"
You can see it here:
People found Musk's use of Sweeney disturbing, while some responded with their own videos, and others called out the cost of AI.
Grok Imagine launched in July 2025. It included a "Spicy" mode that allows users to generate images featuring nudity and sexually explicit content.
As a result, X is under investigation by the California Attorney General's Office and the European Union for the proliferation of nonconsensual sexually explicit content created by Grok Imagine.
In a press release, California Attorney General Rob Bonta said:
"The avalanche of reports detailing the non-consensual, sexually explicit material that xAI has produced and posted online in recent weeks is shocking."
"This material, which depicts women and children in nude and sexually explicit situations, has been used to harass people across the internet."
"I urge xAI to take immediate action to ensure this goes no further."
The European Union released a statement that they were investigating possible X violations of their Digital Services Act.
The EU announced:
"The new investigation will assess whether the company properly assessed and mitigated risks associated with the deployment of Grok's functionalities into X in the EU."
"This includes risks related to the dissemination of illegal content in the EU, such as manipulated sexually explicit images, including content that may amount to child sexual abuse material."
In response, X's official Safety account announced on January 14:
"We will geoblock in jurisdictions where such [AI generated explicit images of real people] content is illegal, the ability of all users in those locations to generate images of real people in bikinis, underwear, and similar attire in Grok on X, and xAI is implementing similar geoblocking measures for the Grok app."
But an investigation by CBS News found users in those jurisdictions were still able to "undress" real people using Grok as late as Monday, January 26.
But at least the videos are longer and the audio is greatly improved.
@annewheaton/Threads
@richardadlof/Threads
@mikeisaphreek/Threads
@daisymouze/Threads