Following a recent spike in sexualized photographs of women and children produced without permission by Elon Musk’s AI chatbot Grok, governments worldwide are taking offense.
On Tuesday, a Polish senator used it as justification to pass regulations about internet safety, while Britain’s top technology official requested that Musk’s social media site X take immediate action.
While politicians and regulators in France, India, Malaysia, and Brazil have blasted the site and demanded investigations, the executive branch of the European Union has slammed Grok.
Growing concern from several countries highlights the terrifying potential of nudification apps that create sexually explicit deepfake images using artificial intelligence.
Creation of images
The issue surfaced following the release of Grok Imagine, an AI image generator that lets users create images and films by entering text prompts, last year. It has an alleged “spicy mode” that can produce adult content.
Grok, which is housed on X, reportedly started approving a lot of user requests to edit other people’s photos late last month, which caused it to blow up.
Grok users may still create pictures of women as of Tuesday by using commands such as “put her in a transparent bikini.”
The issue is made worse by the fact that Grok’s photos are publicly accessible and can thus be widely disseminated, as well as the fact that Musk presents his chatbot as a more daring substitute for competitors with stronger security.
In a report, the nonprofit organization AI Forensics stated that it examined 20,000 photos produced by Grok between December 25 and January 1 and discovered that 2% of them showed someone who looked to be 18 or younger, including 30 young or very young ladies or girls wearing revealing clothing or bikinis.
Musk’s reply
When asked for a reaction, Musk’s artificial intelligence startup, xAI, automatically replied, “Legacy Media Lies.”
X did not, however, contest the existence of the problematic content produced by Grok.
However, it continued to assert in a post on its Safety account that it responds to unlawful content, including images of child sexual abuse, “by removing it, permanently suspending accounts, and working with local governments and law enforcement as necessary.”
Musk’s statement, “Anyone using Grok to make illegal content will suffer the same consequences as if they upload illegal content,” was also repeated by the platform.
An increasing number of nations are calling on Musk to take greater action to control offensive or violent content.
Britain
Technology Secretary Liz Kendall stated on Tuesday that X needs to “urgently” address the issue and that she was in favor of more investigation by Ofcom, the UK’s communications regulator.
The content is “absolutely appalling and unacceptable in a decent society,” according to Kendall.
“We cannot and will not permit the spread of these humiliating and dehumanizing images, which disproportionately target women and girls.”
On Monday, Ofcom announced that it had established “urgent contact” with X.
The watchdog stated, “We are aware of serious concerns raised about a feature on Grok on X that produces undressed images of people and sexualized images of children.”
The watchdog claimed to have gotten in touch with both X and xAI to find out what actions they had taken to adhere to British laws.
Social media companies are required by the UK’s Online Safety Act to stop child sexual abuse and remove any content that they come across.
Poland
On Tuesday, a Polish senator cited Grok as justification for national internet safety legislation that would strengthen safeguards for children and facilitate content removal by authorities.
The speaker of the parliament, Wlodzimierz Czarzasty, stated in an online video that he wished to become Grok’s target to draw attention to the issue and ask the president of Poland to support the legislation.
“Grok has been stripping people lately. It involves undressing toddlers, men, and women. We regret it. He remarked, “To be honest, I would almost want this Grok to undress me too.”
The European Union
According to Thomas Regnier, a spokesman for the European Commission, the bloc’s executive branch is “well aware” that Grok is being used for “explicit sexual content with some output generated with child-like images.”
It’s not spicy. This is against the law. This is terrible. This is repulsive. We perceive it this way, and it has no place in Europe. Grok has already produced comparable performance, he said reporters on Monday.
According to Regnier, the commission requested additional information from Musk’s social media platform X after Grok disseminated content that denied the Holocaust last year. He stated that an analysis of X’s reaction is presently underway.
France
Following complaints from parliamentarians, the Paris prosecutor’s office announced that it is expanding an ongoing probe of X to include sexually explicit deepfakes.
According to a government statement released last week, three government ministers warned prosecutors about “manifestly illegal content” created by Grok and uploaded on X.
The government also raised concerns about potential violations of the EU’s Digital Services Act with the nation’s communications authority.
The government declared, “The internet is neither a lawless zone nor a zone of impunity: sexual offences committed online constitute criminal offences in their own right and fall fully under the law, just as those committed offline.”
India
On Friday, the Indian government sent X an ultimatum, requesting that company remove all “unlawful content” and deal with infringing users.
Additionally, the corporation was instructed by the nation’s Ministry of Electronics and Information Technology to examine Grok’s “technical and governance framework” and submit a report on the steps taken.
By permitting the creation and dissemination of “obscene images or videos of women in a derogatory or vulgar manner to indecently denigrate them,” the ministry accused Grok of “gross misuse” of AI and grave violations of its protections and enforcement.
The ministry cautioned that the corporation would face more serious legal issues if it did not comply with the 72-hour deadline, but India did not provide a public update beyond the deadline.
Malaysia
On Saturday, the Malaysian communications watchdog announced that it was looking into X users who disseminated “grossly offensive, obscene, or indecent content” in violation of the law.
The Malaysian Communications and Multimedia Commission announced that it would call a corporate representative and that it is also looking into online harms on X.
Public concerns about X’s AI capabilities being used to digitally alter “images of women and minors to produce indecent, grossly offensive, or otherwise harmful content” were noted by the watchdog.
Brazilian lawmaker Erika Hilton notified Grok and X to the nation’s data protection agency as well as the federal public prosecutor’s office.
She accused both of creating and disseminating sexualized pictures of women and children without permission in a social media post.
According to her, X’s AI features ought to be turned off until an inquiry is completed.
One of the first transgender senators in Brazil, Hilton, criticized Grok’s ability to digitally modify any public image, “swapping the clothes of women and girls for bikinis or making them suggestive and erotic.”
