Indonesia blocks Grok over non-consensual, sexualized deepfakes

Indonesian officers mentioned Saturday that they’re briefly blocking entry to xAI’s chatbot Grok.

This is likely one of the most aggressive strikes so removed from authorities officers responding to a flood of sexualized, AI-generated imagery — typically depicting actual ladies and minors, and typically showing assault and abuse — posted by Grok in response to requests from customers on the social community X. (X and xAI are a part of the identical firm.)

In a statement shared with the Guardian and different publications, Indonesia’s communications and digital minister Meutya Hafid mentioned, “The federal government views the apply of non-consensual sexual deepfakes as a severe violation of human rights, dignity, and the safety of residents within the digital house.”

The ministry has additionally reportedly summoned X officers to debate the problem.

Diversified governmental responses over the previous week embody an order from India’s IT ministry for xAI to take motion to forestall Grok from producing obscene content material, in addition to an order from the European Fee for the corporate to retain all paperwork associated to Grok, which could possibly be setting the stage for an investigation.

In the UK, the communications regulator Ofcom has mentioned that it’s going to “undertake a swift evaluation to find out whether or not there are potential compliance points that warrant investigation.” Prime Minister Keir Starmer mentioned in an interview Ofcom has his “full assist to take motion.”

And whereas in the US, the Trump administration seems to be staying silent on the problem (xAI CEO Elon Musk is a significant Trump donor and led the administration’s controversial Department of Government Efficiency final yr), Democratic senators have called on Apple and Google to take away X from their app shops.

xAI initially responded by posting a seemingly first-person apology to the Grok account, acknowledging {that a} put up “violated moral requirements and probably US legal guidelines” round youngster sexual abuse materials. It later restricted the AI image-generation feature to paying subscribers on X, although that restriction didn’t seem to have an effect on the Grok app itself, which nonetheless allowed anybody to generate photographs.

In response to a put up questioning why the U.Okay. authorities wasn’t taking motion towards different AI picture era instruments, Musk wrote, “They need any excuse for censorship.”