The chatbot has been under fire over the past two weeks for generating sexually-explicit deepfakes of women and minors.
Elon Musk's AI chatbot Grok is preventing most users from generating or editing any images after a global backlash that erupted after it started spewing sexualised deepfakes, mostly of women.
The chatbot, which can be accessed through Musk's social media platform X, has in the past few weeks been granting a wave of what researchers say are malicious user requests to modify images, including putting women in bikinis or in sexually explicit positions.
Researchers have warned that in a few cases, some images appeared to depict children.
Governments around the world have condemned the platform and opened investigations into the platform.
On Friday, Grok was responding to image altering requests with the message: “Image generation and editing are currently limited to paying subscribers. You can subscribe to unlock these features.”
While subscriber numbers for Grok aren't publicly available, there was a noticeable decline in the number of explicit deepfakes that Grok is now generating compared with days earlier.
Among the official government responses, the European Union has slammed Grok for “illegal” and “appalling” behaviour, while officials in France, India, Malaysia and a Brazilian lawmaker have called for investigations.
On Thursday, Britain's Prime Minister Keir Starmer threatened unspecified action against X.
“This is disgraceful. It’s disgusting. And it’s not to be tolerated,” Starmer said on Greatest Hits radio. “X has got to get a grip of this.”
He said Britain's media regulator Ofcom “has our full support to take action" and that "all options" are on the table.
“It’s disgusting. X need to get their act together and get this material down. We will take action on this because it’s simply not tolerable.”
Ofcom and Britain's privacy regulator both said this week they've contacted X and Musk's artificial intelligence company xAI for information on measures they've taken to comply with British regulations.
Grok launched in 2023. Last summer the company added an image generator feature, Grok Imagine, which included a so-called “spicy mode” allowing users to generate adult content.
The AI chatbot is free to use for X users, who can ask it questions directly on the social media platform. They can either tag it in posts they've created or in replies to posts from other users.
The feature to publicly generate and edit images is currently limited to paying subscribers.
When users attempt to tag Grok and generate images, they are shown a message stating: “Image generation and editing are currently limited to paying subscribers. You can subscribe to unlock these features.”
Image generation and editing, however, remain available on the Grok platform itself.
Charlotte Wilson, the head of enterprise at Check Point, a global cybersecurity firm, said limiting Grok’s image editing to paid users was not a safety fix.
“It may reduce casual misuse, but it will not stop determined offenders, and it does nothing for the victims whose images have already been created, shared, and stored,” Wilson said in a statement.
The issue has been compounded both because Musk pitches his chatbot as an edgier alternative to rivals with more safeguards, and because Grok’s images are publicly visible and can therefore be easily spread across the internet.
"If platforms want to prove they are taking this seriously, they need to treat it like an incident response problem. Close the loophole at the model level so nudification prompts and edits are blocked, not just discouraged,” Wilson added.
