Elon Musk’s platform, X, has announced that its AI tool, Grok, will cease to create sexualized deepfakes following significant public uproar. The tool, which had previously been used to digitally undress images of real individuals, faced intense backlash from both governments and advocacy groups.

In response to the criticism, X has now restricted Grok from altering images to depict people in revealing attire. The company stated, “We have implemented technological measures to prevent the Grok account from allowing the editing of images of real people in revealing clothing such as bikinis.” This new policy applies to all users, including those with paid subscriptions.

This decision follows widespread condemnation over the misuse of Grok to digitally undress women and, alarmingly, even children, without their consent. The move reflects a growing demand for accountability and ethical standards in the use of artificial intelligence.

‘This restriction applies to all users, including paid subscribers.’

It comes amid widespread disgust at a trend that saw Grok being used to remove the clothes of women and even children without their consent.

Many women said they felt violated by the ability of total strangers to create compromising images of them against their will, for all to see.

The UK government, alongside others, piled pressure on Musk to put a stop to the sickening trend, which sparked an urgent debate about online safety laws and AI.

Sir Keir Starmer branded the non-consensual sexual images being produced ‘disgusting’ and ‘shameful’ and media regulator Ofcom launched an investigation.

Grok, the AI chatbot attached to X, will no longer be able to produce sexualised images of real people without their consent

Musk has bowed to pressure from governments and campaigners by limiting his AI tool’s abilities

Part of the update about Grok that was posted to X on Wednesday evening

Last week, the ability to create images with Grok was limited to only users who paid a monthly subscription – but even they will no longer be able to produce scantily-clad edits.

The full climbdown, announced last night, also came hours after California’s top prosecutor said the state was probing the recent spread of AI fakes.

Yesterday, Grok began declining such requests and replying with: ‘Unfortunately I can’t generate that kind of image.’

Starmer welcomed that development in Prime Minister’s Questions, but insisted it did not go far enough.

He insisted the investigation by Ofcom, which has powers to levy fines running into billions of pounds, would still go ahead.

Technology Secretary Liz Kendall said this week she was bringing forward regulations to tighten the law on so-called ‘digital stripping’.

Other countries acted more decisively, with Malaysia and Indonesia blocking Grok altogether amid the fiasco.

The US federal government, meanwhile, refused to condemn Musk’s creation, and Defence Secretary Pete Hegseth said Grok would join Google’s generative AI engine in operating inside the Pentagon network.

Facing MPs at Prime Minister’s Questions on Tuesday, Sir Keir said that more action was needed

After the Prime Minister demanded the US tech billionaire to act to stem the creation of deepfakes the social media tool began refusing to comply with user requests.

The US State Department even threatened the UK, warning that ‘nothing was off the table’ if X was banned.

After the prime minister’s comments at PMQs, Mr Musk wrote on X he was ‘not aware of any naked underage images generated by Grok’ – although the chatbot itself acknowledged it had created sexualised images of children.

Musk continued: ‘Obviously, Grok does not spontaneously generate images, it does so only according to user requests,” he said.

‘When asked to generate images, it will refuse to produce anything illegal, as the operating principle for Grok is to obey the laws of any given country or state.

‘There may be times when adversarial hacking of Grok prompts does something unexpected. If that happens, we fix the bug immediately.’

If X is found to violate the Online Safety Act in the UK, Ofcom can issue a fine of up to ten percent of its global revenue or £18m.

It also has the power to seek to block the site through the courts.

Former Meta boss Sir Nick Clegg has called for tougher regulation of tech firms, branding social media a ‘poisoned chalice’ and the rise of AI online a ‘negative development’.

The ex-deputy prime minister warned that engaging with ‘automated’ content appears to be ‘much worse, particularly for younger people’s mental health’ than interactions with other human beings.