Grok limits image generator after backlash over sexualized AI pictures

Grok, the AI chatbot on Elon Musk’s social media platform X, is limiting image generation and editing capabilities to paying subscribers amid a global outcry from officials and users over the tool being used to create sexualized images of people without their consent. Critics quickly said that merely requiring payment does not address the underlying problems with the tool.

“Image generation and editing are currently limited to paying subscribers. You can subscribe to unlock these features,” Grok posted on X late on Thursday in response to a user, alongside a link to subscribe for premium membership.

A spokesman for Downing Street, the office of the British prime minister, called the restriction to subscribers “insulting to victims of misogyny and sexual abuse” adding that it appeared to turn the feature into a premium service.

The move also drew criticism from victims of nonconsensual AI-generated images.

“It shouldn’t be that you can monetize this. It’s so offensive to all of the victims of this,” said Jess Davies, a British women’s rights campaigner who found herself the victim of being digitally undressed by an X user after she posted a picture of herself on New Year’s Eve.

Davies told The Washington Post that on Friday morning she tested Grok directly through its stand-alone app rather than via X, uploading an image of herself. The tool, she said, generated an image of her in a bikini when prompted. “It even generated fake nipples without me asking,” she said.

Criticism of the chatbot intensified in recent weeks as users noticed that Grok, in response to user prompts, was generating AI images of women in various stages of undress. Users asked the chatbot to show subjects with their clothes removed and replaced with lingerie or bikinis, or to show subjects covered in liquid or with fearful expressions.

The content sparked widespread condemnation from officials, regulators and women’s groups around the world.

In the United Kingdom, communications regulator Ofcom announced this week that it had contacted X following “serious concerns” about a Grok feature that produced images of people undressed and sexualized images of children. In India, the government last week warned X to remove all “unlawful content,” take action against offending users, and review Grok’s “technical and governance framework” within 72 hours.

In Brussels, the European Commission, the European Union’s executive arm, ordered X to retain all internal documents relating to Grok until the end of 2026 while it ensures compliance with its laws regarding illegal content.

Some experts said that X, by making the image generator a premium feature, was not solving the problem but merely profiting from it. “I don’t see this as a victory, because what we really needed was X to take the responsible steps of putting in place the guardrails to ensure that the AI tool couldn’t be used to generate abusive images,” said Clare McGlynn, a law professor at the University of Durham.

McGlynn said she wondered if the decision was driven in part by commercial motives. “Isn’t this a business turning a crisis into another profit-making opportunity to try and encourage people to go on the premium service,” she asked. “That’s possibly why they’ve taken the decision to simply block it for all unpaid users, rather than putting in place the guardrails to prevent it from producing these abusive images.”

Karen Middleton, an expert in social media and marketing at the University of Portsmouth, said that putting the image generator behind a paywall “may reduce casual misuse, but it doesn’t change what the tool can do.”

“Putting ‘nudity’ behind a paywall doesn’t make it safer — it just makes it monetizable,” Middleton said, adding that X should instead focus on “safety-by-design” including disabling the feature and requiring identity and consent checks before transforming images.

X did not immediately reply to a request for comment early Friday, but the company has previously said child sexual abuse material would be removed and the company would work with law enforcement and local governments where necessary to address any violations of law.

Officials in several countries have raised concerns and demanded action. U.K. Prime Minister Keir Starmer called the images “disgraceful” and “not to be tolerated,” in a radio interview Thursday, adding: “X has to get a grip of this.”

In Brussels, European Commission spokesman Thomas Regnier told reporters earlier in the week that officials were aware that Grok offered a “spicy mode” but he said, “This is not spicy, this is illegal.”

Last week, Musk said anyone using the tool to “make illegal content will suffer the same consequences as if they upload illegal content.”

But in a separate post, Musk made light of the controversy, adding two laughing emojis when he re-shared an image of a toaster with a bikini on it, with the original caption: “Grok can put a bikini on everything.”