Know-how Secretary Liz Kendall has referred to as on Elon Musk’s X to urgently take care of its synthetic intelligence chatbot Grok getting used to create non-consensual sexualised photographs of girls and women.
The BBC has seen a number of examples on X of individuals asking the bot to digitally undress individuals to make them seem in bikinis with out their consent, in addition to placing them in sexual conditions.
Kendall mentioned the scenario was “completely appalling”, including “we can not and won’t enable the proliferation of those degrading photographs.”
“It’s completely proper that Ofcom is trying into this as a matter of urgency and it has my full backing to take any enforcement motion it deems needed.”
On Monday, regulator Ofcom said it had made “pressing contact” with Elon Musk’s firm xAI and was investigating considerations Grok has been producing “undressed photographs” of individuals.
X has not responded to any of the BBC’s requests for remark, apart from with an automatically-generated reply stating “legacy media lies”.
However on Sunday, the platform issued a warning to users to not use Grok to generate unlawful content material together with little one sexual abuse materials.
Grok is a free AI assistant – with some paid for premium options – which responds to X customers’ prompts once they tag it in a put up.
It’s usually used to present response or extra context to different posters’ remarks.
However individuals on X are additionally ready to make use of it to edit an uploaded picture by means of its AI picture enhancing function – together with producing what it phrases “spicier” variations of the picture, with out the consent of the particular person depicted.
Girls who’ve stumbled throughout sexualised photographs of themselves made by Grok have described it as dehumanising.
In her assertion Kendall mentioned: “Providers and operators have a transparent obligation to behave appropriately. This isn’t about limiting freedom of speech however upholding the regulation.
“Now we have made intimate picture abuse and cyberflashing precedence offences beneath the On-line Security Act – together with the place photographs are AI-generated. This implies platforms should forestall such content material from showing on-line and act swiftly to take away it if it does.”
Dr Daisy Dixon is among the many feminine X customers who lately began to see individuals take on a regular basis photos she had posted of herself on the platform and ask Grok to undress her or sexualise her.
She instructed the BBC the images left her feeling “shocked”, “humiliated” and frightened for her security.
She added whereas she backed the decision from the Know-how Secretary for motion and located it “heartening”, she nonetheless felt pissed off with X’s lack of accountability.
“Myself and lots of different ladies on X proceed to report the inappropriate AI photographs/movies we’re being despatched each day, however X continues to answer that there was no violation of X guidelines,” she mentioned.
“I simply hope Kendall’s phrases flip into concrete enforcement quickly – I do not wish to open my X app any extra as I am frightened about what I would see.”
