London, Britain – The UK communications regulator Ofcom has announced it is holding “urgent contact” with billionaire Elon Musk’s X (formerly Twitter) and xAI companies over “serious concerns” about the ability of their artificial intelligence tool Grok to generate illegal content including nude images, deepfakes and child sexual abuse material.
Concerns about violating online safety laws
The regulatory body explained in an official statement that it is conducting a “rapid assessment” to determine the platform’s compliance with its legal obligations under the new online safety law, which came into effect last July. The law requires social media companies to prevent and remove child sexual abuse material as soon as it appears and criminalizes the sharing of “fake” intimate images (deepfake) without the subject’s consent.
The most prominent violations observed
Targeting women: Reports from female users about accounts using “Grok” to create nude photos of them without their knowledge.
Child abuse materials: Cases were detected in which the program generated sexual images of minors, according to technical analyses confirmed by Reuters.
Elon Musk’s response and the X platform
For his part, Elon Musk tried to contain the crisis through a post in which he warned that “anyone using Grok to create illegal content will suffer the same legal consequences as if they had uploaded actual illegal content.”
The platform’s official safety account also issued a statement confirming the removal of illegal content, the permanent suspension of involved accounts, and working with governments and law enforcement agencies to prosecute violators.
It also acknowledges that there have been “individual cases” where users have been able to bypass restrictions, while work continues to improve “prevention measures” in xAI technology to prevent such requests altogether.
Potential sanctions
Under British law, Platform X could face formal investigations and hefty fines if it is found to have failed to put in place sufficient controls to prevent artificial intelligence from generating child pornography or content that violates users’ physical privacy.
