"Abhorrent": Great Britain to take action against Grok's sexualized deepfakes
In Great Britain, apps that digitally undress people are to be banned. A new law will be enforced against the practice on Grok.
(Image: Photo Agency/Shutterstock.com)
In light of the ongoing scandal surrounding sexualized deepfakes from the AI chatbot Grok, the British government has announced that it will begin enforcing a law against it this week and also intends to ban applications that can be used to digitally undress people. Technology Minister Elizabeth Kendall said this on Monday in the British House of Commons, sharply criticizing X's actions. Previously, the British media regulator had presented the investigations against the microblogging service, which had already become public, and pointed out that they could result in severe penalties.
High penalties already possible
Kendall described the content generated with the help of the AI chatbot as “abhorrent” and an “affront to decent society.” Furthermore, they are illegal. X's decision to restrict image generation to paid accounts on Friday was “not nearly enough.” It is an insult to the victims to say that the service remains accessible; you just have to pay for it. A law from last year criminalizes the generation or request for generation of intimate images without the consent of the affected parties. Enforcement will begin this week. In addition, a currently prepared law will explicitly criminalize apps that provide AI-generated nude photos.
Previously, the media regulator Ofcom published details of the investigations against X, which also concern the sexualized deepfakes that Grok has publicly generated in large quantities. It is being investigated whether the microblogging service is taking appropriate steps to ensure that users on X do not see depictions of child abuse and how the service minimizes risks. In addition, it is being examined how the service ensures that illegal content is removed from the platform as quickly as possible. The investigations are based on a response from X to a corresponding inquiry. Should violations be found, a penalty of 18 million pounds (over 20 million euros) or 10 percent of global turnover could be imposed.
Videos by heise
Grok's sexualized deepfakes have been causing outrage worldwide for days. Initially, users could simply ask for any photo on the platform to be depicted, for example, in a bikini. The AI chatbot then complied with this or similar requests, sometimes even for minors. On Friday, image generation was restricted to such an extent that it could only be requested directly on X by paid accounts. However, Grok continues to fulfill such requests in direct messages or on its website, even for non-paying users. As the first countries, Indonesia and Malaysia have therefore had the chatbot completely blocked over the weekend.
(mho)
