Elon Musk’s Grok bot restricts sexual image generation after outcry

Elon Musk’s Grok chatbot on Thursday began limiting the power for non-paying customers to create deepfake, sexualized photographs after international pushback from customers and governments.

Late final month, some customers started tagging Grok on Musk’s social media platform X with prompts akin to “put her in a bikini” to generate nonconsensual photographs of actual individuals on X.

The morphed photographs, together with these of celebrities, politicians, and a few minors, had been created by Grok and posted publicly on X.

Earlier than the present restrictions kicked in, Musk participated within the AI undressing development by instructing Grok to create photographs of himself in a bikini.

Customers and authorities complained and threatened to sue or punish X. Grok initially apologized for the inappropriate photographs of what appeared like kids.

Now X has put Grok’s controversial morphing capabilities behind a paywall.

Governments, together with the UK and the European Fee, stated they aren’t happy with X’s safeguards as the power to create and edit photographs continues to be available to paying customers.

“Whereas limiting AI picture era to paid customers might marginally cut back quantity and enhance traceability, the abuse has not been stopped. It has merely been positioned behind a paywall, permitting X to revenue from hurt,” stated Emma Pickering, head of technology-facilitated abuse and financial empowerment at Refuge, in an announcement.

As the general public fury over what many have referred to as AI-enabled harassment grows, regulators have taken discover, with some proposals to ban X of their nations.

Prime Minister of the U.Ok., Keir Starmer, stated that X has “bought to get a grip” of this “disgusting” development. “We’ll take motion on this as a result of it’s merely not tolerable,” Starmer stated.

In a veiled pushback, Musk appeared to criticize U.Ok. regulators outrage by reposting a declare that ChatGPT additionally permitted the creation of a bikini picture of a non-real individual by way of artistic, non-explicit prompting.

After condemning Grok for producing sexualized photographs, the European Fee ordered X to retain all paperwork referring to Grok to make sure compliance with its guidelines.

“We’d like to have the ability to have entry to them if we request it explicitly,” a spokesperson for The Fee stated in a news conference.

France, India and Malaysia, and Brazil are additionally scrutinizing the platform.

Whereas public figures had been early targets for AI undressing, it quickly tipped over to personal people. On January 3, one person prompted Grok 50 instances a day to generate nonconsensual, sexualized photographs of girls in office settings, in line with an evaluation by AI-detection firm Copyleaks.

Gender justice group Ultraviolet has mounted a campaign to stress Apple and Google to take away X from their respective app shops on the grounds that X violates their little one sexual abuse pointers.

Musk has positioned his chatbot as a extra edgy different in comparison with options akin to Gemini, ChatGPT and Claude, which have extra restrictive guidelines round express content material requests.