Grok controversy: Why Elon Musk’s X backed down, restricted AI tool from creating sexualised images

In several instances, users replied to photos of women, often taken from public posts, and prompted Grok to digitally undress them or portray them in suggestive poses. The AI bot promptly complied, only stopping short of full nudification which the site does not allow.

The AI-generated images appeared publicly in the same threads, exposing the targeted persons to harassment. There were also several instances of the tool being used to create such images of children.

This raised serious concerns over the misuse of AI and the lack of guardrails on the platform. 

How did X and Musk respond?

After global outcry at the harmful nature of the content, Musk posted on January 3 that “anyone using Grok to make illegal content will suffer the same consequences as if they upload illegal content”.

Last week, the company limited such image generation capabilities only to paid users.

On January 14, Musk denied knowledge of the AI service being to create such images of children. He posted: “I am not aware of any naked underage images generated by Grok. Literally zero.” 

“Obviously, Grok does not spontaneously generate images, it does so only according to user requests. 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,” he said. 

He said an “adversarial” hacking of Grok may lead to the service doing something “unexpected,” but the company fixes such bugs “immediately”. 

Just hours after his comments, the company said it has restricted the service to stop generating such images altogether, no matter whether the prompt is made by a paying user of X or not. 

What prompted this backdown?

Last week, after a stern notice by the Indian government, X removed 3,500 pieces of content and blocked 600 accounts while admitting its mistake. The government had expressed dissatisfaction over X’s response to its January 2 notice on its failure to observe due diligence obligations under the Information Technology Act, 2000 and associated rules. 

The platform faced a similar backlash in the UK, Malaysia and Indonesia. In the UK, the law is set to change this week to criminalise the creation of such images. Malaysia and Indonesia have blocked access to Grok and are pursuing legal action against X and Musk’s AI unit xAI, alleging failures to prevent harmful content and protect users.

The California Attorney General also announced that he was opening an investigation into Grok and xAI over the images.

Hours later, the company said in a post on X: “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 restriction applies to all users, including paid subscribers.”

It added that image creation and the ability to edit images via the Grok account on the X platform are now only available to paid subscribers. The company would also geoblock the ability of all users to generate images of real people in bikinis, underwear, and similar attire via the Grok account and in Grok in X in jurisdictions where it’s illegal.

“We remain committed to making X a safe platform for everyone and continue to have zero tolerance for any forms of child sexual exploitation, non-consensual nudity, and unwanted sexual content,” the company said. 

The scrutiny in India 

As reported by The Indian Express last week, the social media platform had told the IT Ministry that it was open to permanently disabling accounts which engaged in such activity on the platform and committed to taking action against accounts creating such inappropriate prompts on its artificial intelligence platform. The platform had already blocked pieces of such content that were flagged by the government. 

The government, though, was unconvinced by the company’s response. There was no technical explanation and no mention of steps to prevent Grok from generating such images in the first place, The Indian Express had learnt. It was also looking at legal options to challenge the platform. 

In its formal response, sent on Wednesday, X had acknowledged the severity of the issue and had committed to taking action against accounts creating such inappropriate prompts on its artificial intelligence platform. The company had assured the government that it will strictly comply with India’s Information Technology Act, 2000 and associated rules governing digital content.