Elon Musk's AI chatbot Grok has started restricting free users from generating or editing images after a global backlash over sexualized deepfakes of individuals, but the change has not satisfied authorities in Europe.
In recent weeks, the chatbot, accessed through Musk's X social network, has been inundated with malicious requests from users to modify images, such as putting women in bikinis or sexually explicit positions, according to researchers.
Researchers have warned that in some cases, some images appeared to depict children. Governments around the world have condemned the platform and launched investigations.
On Friday, Grok responded to requests to alter images with the message: "Image generation and editing are currently limited to paid subscribers. You can subscribe to unlock these features."
While Grok's subscriber numbers are not publicly available, on Friday there was a notable decrease in the number of explicit deepfakes Grok was generating, compared to a few days earlier.
The chatbot was still handling image requests, but only from X users with blue verification marks granted to premium subscribers paying $8 per month for features that include higher usage limits for the chatbot.
So far, an X spokesperson has not responded to a request for comment.
The restrictions for non-paying users did not appear to change the views of leaders or regulators in Europe.
"This doesn't change our fundamental problem. Paid subscription or not, we don't want to see such images. It's as simple as that," said Thomas Regnier, a spokesperson for the European Commission's Executive. The Commission had already criticized Grok for "illegal" and "abhorrent" behavior.
The British government was also not satisfied.
Grok's changes "are not a solution," said Geraint Ellis, a spokesperson for UK Prime Minister Keir Starmer, who had threatened on Thursday to take unspecified action against X.
"In fact, it's insulting to the victims of misogyny and sexual violence," he stated, noting that this shows X "can act quickly when it wants to."
"We expect swift action," he said, adding that "all options are on the table."
In a statement to Greatest Hits radio, Starmer said X needs to "get a grip and remove this material. We will take action on this because it's simply not acceptable."
UK media and privacy regulators said this week they have contacted X and Musk's AI company xAI to inquire about measures taken to comply with British regulations.
France, Malaysia, and India have also scrutinized the platform, and a Brazilian lawmaker has called for an investigation. The European Commission has ordered X to preserve all internal documents and data related to Grok until the end of 2026, as part of a broader investigation under the EU's Digital Services Act.
Grok is free for X users, who can ask it questions on the social network. They can tag it in posts they have created directly or in responses to other users' posts.
Grok was launched in 2023. Last summer, the company added an image generator feature, Grok Imagine, which included a so-called "spicy mode" that could generate adult content.
The issue is compounded by the fact that Musk presents his chatbot as a more daring alternative to rivals with more safeguards, and because Grok's images are publicly visible, making them easily shareable.
