Grok, the AI assistant on X, has been making headlines this week after it repeatedly generated explicit deepfakes of underage women and girls. This prompted both Ofcom (the UK's independent regulator for the communications industries) and the government's Technology Secretary to intervene.
"We are aware of serious concerns raised about a feature on Grok on X that produces undressed images of people and sexualised images of children," Ofcom said in a statement, adding that it has made "urgent contact" to ensure "legal duties" are complied with. "We will undertake a swift assessment to determine whether there are potential compliance issues that warrant investigation," it continued.
Just one day after Ofcom confirmed it is looking into concerns about Grok, MP Liz Kendall called on Elon Musk, who owns X, to urgently deal with Grok and its widespread misuse. Describing it as "absolutely appalling", she said: "We cannot and will not allow the proliferation of these degrading images."
"It is absolutely right that Ofcom is looking into this as a matter of urgency," the Technology Secretary continued. "It has my full backing to take any enforcement action it deems necessary."
Here's what you need to know about Grok, why the AI assistant is causing so much controversy, and what X is doing about it...
What is Grok?
Grok is an AI assistant on X. It was developed by xAI, Elon Musk's artificial intelligence company, and was first launched on the social media platform (formerly known as Twitter) in 2023. Grok is free, although there are some paid-for premium features, which we'll explain in a moment.
Everyone's clicking on...
What does Grok do?
Grok was designed to respond to prompts from X users and give reactions or provide more context on posts. It can also generate AI images and videos, if requested, using its Imagine feature.
"Grok is an AI assistant who helps complete tasks, like answering questions, solving problems, and brainstorming," X says of Grok's purpose. Ultimately, X says, "Grok is here to help you with your tasks while keeping you entertained and engaged."
Unlike some other AI models, Grok has an in-built personality. "Grok is designed to answer almost any question with a touch of wit and humour, while also providing helpful and insightful responses," explains X. "Grok has a rebellious streak and an outside perspective on humanity, making it a unique and entertaining companion."
What is Grok's Spicy Mode?
Grok has a Spicy Mode within its Imagine feature, which allows users to create suggestive and partially nude content that is typically restricted on other AI tools. Grok's Spicy Mode is not free to use and is part of paid subscriptions, either to X's Premium+ or SuperGrok.
Why is there controversy over Grok?
Social media users have been using Grok's Spicy Mode to make non-consensual, explicit images of women and girls. This has, rightfully, prompted a lot of backlash, leading to Ofcom and the UK Government's Technology Secretary stepping in.
Charities, like Refuge, have also shared concerns about Grok. "The disturbing rise in AI intimate image abuse, facilitated by platforms such as Grok, is not just a digital threat – it has dangerous consequences for women and girls," said Emma Pickering, Head of Technology-Facilitated Abuse and Economic Empowerment at Refuge.
"Generative AI has made it easier than ever for perpetrators to create fake images at the expense of women's safety, and at Refuge, we see firsthand the long-term impact that all forms of intimate image abuse can have on a survivor's mental health and wellbeing," Pickering added.
As for how the issue should be tackled, she suggested that tech companies need to be held accountable for "implementing effective safeguards and preventing perpetrators from causing harm."
However, while a law designed to criminalise the sharing of non-consensual deepfakes has been making its way through Parliament (and sharing real images is already illegal), it's yet to come into effect and many of the platforms that host such content are based overseas – leaving the government in a difficult position when it comes to properly holding them to account.
Add to this, President Trump recently put a multi-billion-pound 'tech prosperity' deal between the US and UK on ice, leaving Starmer and co in a murky middle spot of wanting to appease the President and the tech bros of his inner circle, while pledging to tackle violence against women and girls online.
"As technology evolves, women and girls' safety depends on tighter regulation around image-based abuse, whether real or deepfake, as well as specialist training for prosecutors and police," Pickering added. "Women have the right to use technology without fear of abuse, and when that right is violated, survivors must be able to access swift justice and robust protections."
How did X respond to the misuse of Grok?
In response to multiple flags and queries from social media users about the misuse of images, Grok itself said on X: "I've reviewed recent interactions. There are isolated cases where users prompted for and received AI images depicting minors in minimal clothing."
It continued, "xAI has safeguards, but improvements are ongoing to block such requests entirely."
Meanwhile, X said in a statement on the platform: "We take action against illegal content on X, including Child Sexual Abuse Material (CSAM), by removing it, permanently suspending accounts, and working with local governments and law enforcement as necessary."
Musk has also commented on the matter, albeit briefly. "Anyone using Grok to make illegal content will suffer the same consequences as if they upload illegal content," he said on X in response to someone who said, "Some people are saying Grok is creating inappropriate images. But that's like blaming a pen for writing something bad. A pen doesn't decide what gets written. The person holding it does. Grok works the same way. What you get depends a lot on what you put in."
What are the new changes to X's Grok?
X has confirmed that image editing on Grok via the platform will now be limited to paid subscribers. Previously, only Spicy Mode image-editing was paid for, but this has now been extended to all image editing on Grok. Those who do not subscribe to X, however, can still use Grok to edit images on its separate app and website.
As of this morning (9 January), Grok now tells users asking it to alter images uploaded to X that "image generation and editing are currently limited to paying subscribers".
"[Users] can subscribe to unlock these features," the chatbot adds.
Unfortunately, industry experts aren't confident that the change will have a positive impact on deepfakes being made. "The recent decision to restrict access to paying subscribers is not only inadequate – it represents the monetisation of abuse," said Pickering at Refuge.
"While limiting AI image generation to paid users may marginally reduce volume and improve traceability, the abuse has not been stopped," she continued. "It has simply been placed behind a paywall, allowing X to profit from harm. This response comes far too late, after a disturbing number of images have already been created and circulated. Traceability offers little comfort when conviction rates for intimate image abuse remain woefully low."
Cosmopolitan UK has reached out to xAI for comment.
Jennifer Savin is Cosmopolitan UK's multiple award-winning Features Editor, who was crowned Digital Journalist of the Year for her work tackling the issues most important to young women. She regularly covers breaking news, cultural trends, health, the royals and more, using her esteemed connections to access the best experts along the way. She's grilled everyone from high-profile politicians to A-list celebrities, and has sensitively interviewed hundreds of people about their real life stories. In addition to this, Jennifer is widely known for her own undercover investigations and campaign work, which includes successfully petitioning the government for change around topics like abortion rights and image-based sexual abuse. Jennifer is also a published author, documentary consultant (helping to create BBC’s Deepfake Porn: Could You Be Next?) and a patron for Y.E.S. (a youth services charity). Alongside Cosmopolitan, Jennifer has written for The Times, Women’s Health, ELLE and numerous other publications, appeared on podcasts, and spoken on (and hosted) panels for the Women of the World Festival, the University of Manchester and more. In her spare time, Jennifer is a big fan of lipstick, leopard print and over-ordering at dinner. Follow Jennifer on Instagram, X or LinkedIn.
