“I was involuntarily undressed by Elon Musk’s Grok on X,” Swedish Deputy Prime Minister Ebba Busch said in a recent video on the platform. She is one of countless victims of a feature on X’s AI chatbot, Grok, that allows users to digitally strip women and children of their clothing. In less than two weeks, Grok generated three million sexualized images, including tens of thousands that appear to depict children.

Now, the European Commission, the EU’s executive arm, has launched an investigation into Grok under the Digital Services Act, or DSA. The Commission said it would “assess whether the company [X] properly assessed and mitigated risks associated with the deployment of Grok’s functionalities into X in the EU.”

The investigation could set an important precedent in confronting US tech firms and protecting citizens from privacy violations. “If they can go up against Musk, that is huge,” Joanna Bryson, professor of ethics and technology at the Hertie School university in Berlin, told DW. “It tells everybody the EU is serious.”

EU has the laws to protect from AI harm

But how much power does the EU have? Under its laws, the bloc can regulate digital services much like physical products. Just as unsafe toys or appliances can be banned from the European market, online platforms that break EU rules can be forced to change — or be shut out entirely.

“The DSA is arguably the most powerful platform regulation in the world,” said Philipp Hacker, professor of law and ethics at the Viadrina University in Frankfurt an der Oder. Rather than targeting specific technologies, it addresses broad categories such as “systemic risks,” giving the European Commission flexibility to respond to new harms as AI evolves. Grok’s undressing function did not exist when the DSA entered into force in 2022.

The EU has two main enforcement tools. It can fine X up to 6% of its global annual turnover, a measure it already used in December, when it fined the platform €120 million ($143 million) for breaching DSA transparency rules. In cases of repeated violations, the Commission can also block the platform entirely within the EU. For now, a fine is seen as the more likely next step.

The EU’s leverage is strengthened by its market size. As one of the world’s largest consumer markets, it wields significant influence over global tech firms.

EU reportedly delayed start of Grok investigation

Still, experts question whether the Commission will push hard enough to set a lasting precedent. “We have a lot of provisions protecting fundamental rights. The problem is there enforcement, making sure they are not disregarded,” says Marco Bassini, assistant professor of Fundamental Rights and Articifical Intelligence at Tilburg University.

At a time of strained EU-US relations, Brussels has been cautious about fully deploying its regulatory power against American tech companies. According to reporting by the German newspaper Handelsblatt, the launch of the Grok investigation was delayed by the office of Commission President Ursula von der Leyen amid a tariff dispute between the EU and the US over Greenland. In the past, US Commerce Secretary Howard Lutnick linked tariff reductions to weakening EU tech regulation.

“I think a big mistake was made,” Member of the European Parliament (MEP) Alexandra Geese told DW. “There was a lot of fear about the reaction from the US administration.”

Experts doubt how far the EU is willing to go
While the eventual launch of the investigation is seen as a positive signal, doubts remain over how far the Commission is willing to go. “One has to distinguish between the legal position and the geopolitical position,” said Bryson. “The Commission looks at both.”

There is also skepticism over whether a 6% fine would be enough to change the behavior of the world’s wealthiest man. “We should increase this number so companies really pay attention to what the legislature expects in Europe,” MEP Axel Voss told DW.

There is no concrete timeline on the investigation. According to experts, it could take months for the Commission to reach a result. (DW)

The post Can the EU take on Musk over AI “undressing” feature? appeared first on Newswire.

Elon Musk’s xAI has announced it will block the ability of its Grok AI tool to alter images of real people to put them in “revealing clothing such as bikinis”, amid a global backlash over the tool being used to generate explicit imagery.

The move came just hours after the billionaire said he was was not aware of any “naked underage images” made by Grok.

Scrutiny of Grok has intensified worldwide, with some people using it to digitally undress women and children without their consent and posting those images to X. Thousands of these sexualized AI images have inundated X over the last few weeks. xAI, which makes Grok, also owns X – both are run by Musk.

“This adds an extra layer of protection by helping to ensure that individuals who attempt to abuse the Grok account to violate the law or our policies can be held accountable,” the company posted on X on Wednesday evening. It also said that it has “zero tolerance for any forms of child sexual exploitation, non-consensual nudity, and unwanted sexual content”.

The changes X announced on Wednesday will apply to all users on the social media platform, even paying subscribers, the company said. X did not specify, however, whether people would still be able to create such images on the standalone Grok app.

Earlier on Wednesday, Musk wrote in a post on X: “I [sic] not aware of any naked underage images generated by Grok. Literally zero.” Musk’s xAI and X have faced growing backlash globally, including an investigation by California’s attorney general and calls by lawmakers and advocacy groups for Apple and Google to drop Grok from app stores. UK regulators have also launched an investigation, and countries including Malaysia and Indonesia have instigated bans or legal action .

Musk has said that Grok was programmed to refuse illegal requests and must comply with the laws of any given country or state.

“Obviously, Grok does not spontaneously generate images. It does so only according to user requests,” Musk said on Wednesday.

Musk has said earlier on X that anyone using Grok to make illegal content would suffer the same consequences as if they uploaded illegal content.

Three Democratic US senators last week called on Apple and Alphabet’s Google to remove X and its built-in AI tool Grok from their app stores, citing the spread of nonconsensual sexual images of women and minors on the platform.

A coalition of women’s groups, tech watchdogs and progressive activists also called on the tech giants for a similar move.

Last week, X curtailed Grok’s ability to generate or edit images publicly for users who weren’t paying subscribers. However, industry experts and watchdogs have said that Grok was still able to produce sexually explicit images, and that restrictions, such as paywalling certain features, may not fully block access to deeper AI image tools.

In the UK, the law is set to change this week to criminalize the creation of such images, and the prime minister, Keir Starmer, said on Wednesday that X was working to comply with the new rules. Communications regulator Ofcom is investigating the AI tool.

X and xAI did not return request for comment. (The Guardian)

The post Musk’s X to block Grok AI from creating sexualized images of real people appeared first on Newswire.