Technology's spotlight should be on abusers, not victims
22 Feb 2026
The proliferation of intrusive and humiliating online representations of female celebrities and others serves as a reminder that more needs to be done with regards to data protection and regulation of internet technology, says Professor Brian J Ford…
The sign read ‘Scientific Restoration of Any Photo’ so she laid a picture on the counter. “This is my favourite photograph of my late husband,” she said. “But he is wearing that awful fishing hat. Could you retouch the picture to get rid of it?”
The technician smiled at her. “Of course we can,” he said. “What kind of hair did your husband have?”
“Why ask me?” she retorted. “You’ll see that when you take off his hat.”
Harmless enough: much more serious when it’s not a hat you’re removing, but clothes. Grok is hitting the headlines for facilitating the creation of nude images of unwitting individuals. That’s invasive, intrusive and traumatic. It’s also unreal – all an AI can do is superimpose an assumed nude appearance. It’s not real. And it’s not new.
There is a Roman erotic, nude image of Cleopatra dating from about 30 BC showing her astride a crocodile to exemplify her supposed sexual appetite. During the following century there were erotic naked portrayals of Stateira, the Persian wife of Alexander the Great and numerous pictures euphemistically entitled Venus which were actually portrayals of people including Livia Drusilla, the wife of Emperor Augustus.
Renaissance art often emphasised eroticism in the name of art; Napoleon’s sister Pauline, Princess Borghese, was imagined semi-nude in a marble statue, and a titillating Marie Antoinette was painted by Vigée-Le Brun. Some – like Victorine Meurent, who posed naked for Eduard Manet’s picture Olympia – were nude models paid to pose, but most portrayals were semi-naked, sometimes erotic, images of women who’d have been horrified to see the result.
The fact you can create an anonymous presence online is the greatest problem with social media. People who post need to be accountable. And those who publish degrading images need sanction
Non-consensual nude images featured in Hustler magazine’s ‘Beaver Hunt’ in 1980, which published private photos submitted maliciously, and spreading non-consensual images became headline news in 2008 when highly intimate photos of a Canadian-born actor, Edison Chen, were stolen and distributed. A number of prominent women featured in the images, many of them ostensibly happily married and some promoting a chaste lifestyle. Then in 2014 the iCloud was hacked, and images of Jennifer Lawrence, Kate Upton and others brought non-consensual celebrity nudes into public awareness.
As computer science advanced and photo-editing improved, creating indecent images of celebrities became available to all. The first public examples were pornographic videos posted by a Reddit user named “deepfakes” in 2017 who first applied face-swap tools to existing adult videos. A short-lived app called DeepNude followed in June 2019, and it was to spawn many others.
Grok was a late arrival on the scene. Users discovered they could reply to almost any public photo of women, celebrities, and even minors with prompts like ‘put her in a bikini’, ‘make her topless’, and far worse. It caught on, and there were soon over 6,000 bikini-related edits each hour. Violent obscene videos soon appeared. Elon Musk and his xAI team downplayed the problem as “legacy media fabrication” so it mushroomed for a good ten days. Grok was built to be permissive and integrated with posts on but this policy had spectacularly backfired. Eventually xAI restricted image manipulation to paid subscribers only, in the vague hope of making abusers more traceable. Most other social media won’t permit anything like that.
In Britain, section 138 of the Data (Use and Access) Act 2025 makes it a crime to create non-consensual intimate images of people, even if they are just for you to see. Grok has meanwhile been banned in Malaysia and Indonesia, the UK and USA are planning legal action, and the French authorities have raided the Paris offices of xAI and summoned Musk to account for his behaviour.
Governments everywhere are considering controls on social media for the young: but what about the old? Much of what is published online is fake; yet there is no redress. I have always said that the fact you can create an anonymous presence online is the greatest problem with social media. People who post need to be accountable. And those who publish degrading images need sanction. They cause ineradicable harm to their victims, and that, surely, must end.
Pic: Shutterstock (Davide Angelini)
