Photo Credit: Alexander Shatov
Elon Musk’s AI chatbot now prevents non-paying users from generating or editing images in the wake of global backlash over its sexualized deepfakes.
Grok, Elon Musk’s AI chatbot on X, the platform formerly known as Twitter, no longer allows non-paying users to generate or edit images. The move comes amid intense backlash on a worldwide scale against sexualized deepfakes of people, particularly minors.
Users on X have been using Grok to generate a wave of “malicious user requests” to modify images by feeding photos—such as Taylor Swift at a red carpet event, or a woman in tight clothing—to the chatbot and asking it to remove the clothing in its output. Reports caution that in some cases, sexualized images appeared to depict children. As a result, governments across the globe have condemned the platform, while some have opened investigations.
Now, Grok is responding to image-altering requests from non-paying users with a message stating that such features “are currently limited to paying subscribers. You can subscribe to unlock these features.”
Researchers note Grok was still granting such image requests, but only from premium subscribers—those with blue checkmarks who pay $8 a month for access to features including higher Grok usage limits. It’s also worth mentioning that the restrictions for what paying subscribers can generate on Grok do not appear to have changed. The Associated Press also confirmed on Friday that Grok is still fully available for free users on the standalone Grok website and app.
The restrictions for non-paying users did not impress regulators in Europe, however. Thomas Regnier, a spokesperson for the European Union’s Executive Commission, said the fundamental issue had not changed. The Commission had earlier called the platform’s behavior “illegal” and “appalling.”
“Paid subscription or non-paid subscription, we don’t want to see such images. It’s as simple as that,” said Regnier.
Geraint Ellis, a spokesperson for British Prime Minister Keir Starmer, also stated that the Grok changes were “not a solution.”
“It is insulting to the victims of misogyny and sexual violence,” said Ellis, remarking that Musk’s platform “can move swiftly when it wants to do so,” and the changes to Grok only reinforce that fact. “We expect rapid action.”
Notably, Starmer threatened unspecified legal action against the former Twitter on Thursday, saying that they needed to “get their act together and get this material down.”
Unauthorized sexually explicit deepfake content has skyrocketed since the dawn of genAI, and Grok is far from the only offender—but researchers say it is by far the biggest offender.
Since Grok’s launch in 2023, Musk has pitched the chatbot as an “edgier alternative” to its competitors. The bot’s image generator, which launched last summer, includes a “spicy mode” that can output adult content. Grok’s images are also publicly visible, which can therefore spread deepfakes like wildfire.
