Zürcher Nachrichten - Musk's Grok under fire over sexualized images despite new limits

Grok has faced global backlash after it emerged the feature allowed users to sexualize images of women and children using simple text prompts such as "put her in a bikini" or "remove her clothes."

Grok appeared to deflect the criticism with a new monetization policy, posting on the platform X late Thursday that image generation and editing were now "limited to paying subscribers," alongside a link to a premium subscription.

British Prime Minister Keir Starmer's office joined the chorus of critics, condemning the move as an affront to victims and "not a solution."

"That simply turns an AI feature that allows the creation of unlawful images into a premium service," a Downing Street spokesperson said.

"It's insulting the victims of misogyny and sexual violence."

EU digital affairs spokesman Thomas Regnier said "this doesn't change our fundamental issue, paid subscription or non-paid subscription. We don't want to see such images. It's as simple as that."

"What we're asking platforms to do is to make sure that their design, that their systems do not allow the generation of such illegal content," he told reporters.

The European Commission, which acts as the EU's digital watchdog, has ordered X to retain all internal documents and data related to Grok until the end of 2026 in response to the uproar.

- 'Safety gaps' -

Grok, developed by Musk's startup xAI and integrated into X, announced the move after Wednesday's fatal shooting in Minneapolis by an immigration agent, which triggered a wave of AI deepfakes.

Some X users used Grok to digitally undress an old photo of the victim, as well as a new photo of her body slumped over after the shooting, generating AI images showing her in a bikini.

Another woman wrongly identified as the victim was also subjected to similar manipulation.

The fabricated images still appeared to float around X -- and spread to other tech platforms -- on Friday despite the new restriction.

There was no immediate comment from X on the Minneapolis deepfakes.

When reached by AFP for comment by email, xAI replied with a terse, automated response: "Legacy Media Lies."

"Restricting Grok's image-generation tools to paying subscribers may help limit scale and curb some misuse, but it doesn't fully address the safety gaps that allowed nonconsensual and sexualized content to emerge," said Cliff Steinhauer, from the nonprofit National Cybersecurity Alliance.

"Access restrictions alone aren't a comprehensive safeguard, as motivated bad actors may still find ways around them, and meaningful user protection ultimately needs to be grounded in how these tools are designed and governed."

France, Malaysia and India have also previously pushed back against the use of Grok to alter women and children's photos, after a flood of user complaints, announcing investigations or calling on Musk's company for swift takedowns of the explicit images.

Britain's communications regulator Ofcom announced earlier this week that it had made "urgent contact with X and xAI" over the Grok feature, warning that it could open an investigation depending on their response.

On Friday, an Ofcom spokesperson said the regulator had "received a response" and was now "undertaking an expedited assessment as a matter of urgency."

Last week, in response to a post about the explicit images, Musk said that anyone using Grok to "make illegal content will suffer the same consequences as if they upload illegal content."

But he appeared to make light of the controversy in a separate post, adding laughing emojis as he reshared to his 232 million followers on X a post featuring a toaster wrapped in a bikini.

"Grok can put a bikini on everything," the original post said.

burs-lga-raz-ac/aha

O.Meier--NZN