X Continues to Allow Posting of Sexualised AI-Generated Images Despite Restrictions

X, formerly known as Twitter, has been found to still permit users to post highly sexualised videos of women in bikinis generated by its AI tool Grok, despite recent claims of stricter controls. The Guardian demonstrated that it was possible to create short videos depicting women stripping to bikinis from photos of fully clothed individuals using Grok. These videos could then be shared publicly on X without any visible moderation, making them accessible to any user within seconds. This loophole seemingly bypasses the new restrictions announced by X earlier in the week, which had been praised by UK Prime Minister Keir Starmer, who condemned the AI-generated images as "disgusting" and "shameful."

In response to growing public concern, X stated it had implemented technological measures to prevent Grok from editing images of real people in revealing clothing, such as bikinis, and asserted a zero-tolerance policy towards child sexual exploitation, nonconsensual nudity, and unwanted sexual content. However, the company did not clarify whether such content could still be generated using the standalone Grok Imagine app, accessible via web browsers. Investigations revealed that Grok Imagine continues to respond to prompts that digitally remove clothing from images of real women, even creating short videos of provocative striptease acts. This indicates that while restrictions may apply within X's platform, the standalone app remains a conduit for producing and sharing sexualised AI-generated content.

Advocates and experts have voiced strong criticism of X's handling of the issue. Rebecca Hitchen from the End Violence Against Women Coalition emphasized that the ease of access to such nudification tools shows a lack of seriousness from X regarding online violence against women and girls. She urged UK regulators and the government to exert pressure on X and other platforms to curb the spread of image-based sexual abuse. Similarly, Penny East, CEO of the Fawcett Society, highlighted the inconsistency in Musk's approach, noting that despite promises to restrict nudification, the problem persists. She criticized the tech sector's failure to prioritize user safety and dignity, underscoring the unacceptable reality that women cannot expect to engage online without being subjected to digital undressing.

Government officials have expressed cautious approval of X's announced measures but remain vigilant. Technology Secretary Liz Kendall described the sexual manipulation of images as "despicable and abhorrent" and welcomed the steps taken, while emphasizing the need for a thorough investigation by Ofcom, the UK media regulator. Prime Minister Keir Starmer demanded immediate compliance with UK law, stressing that free speech does not extend to violating consent. Ofcom confirmed its ongoing investigation into X, working intensively to understand the failures and ensure corrective action. Internationally, Canada's privacy watchdog is investigating xAI, the company behind Grok, while authorities in the Philippines and Malaysia are considering blocking or taking legal action against the tool.

Interestingly, the controversy has seemingly boosted public awareness of Grok. Elon Musk shared a post noting a surge in Grok's popularity and real-world usage globally, encouraging users to try the tool. Meanwhile, the UK government reiterated that the Online Safety Act mandates platforms like X to prevent illegal content, including nonconsensual intimate images and child abuse material. Further legislative efforts are underway, such as the proposed "nudification" offence targeting tools designed to generate nonconsensual intimate images. This ongoing situation highlights the complex challenges of regulating AI-generated content, balancing technological innovation with user safety and legal compliance.