California AG sends Musk’s xAI a cease-and-desist order over sexual deepfakes

Earlier this week, the California Attorney General’s bureau announced that it was investigating xAI implicit reports that the startup’s chatbot, Grok, was being used to make nonconsensual intersexual imagery of women and minors. On Friday, the authorities followed up by sending a cease-and-desist missive to the company, demanding that it instrumentality contiguous enactment to halt the accumulation of nonconsensual intimate images and CSAM — kid intersexual maltreatment material.

“Today, I sent xAI a cease-and-desist letter, demanding the institution immediately stop the instauration and organisation of deepfake, nonconsensual, intimate images and kid intersexual maltreatment material,” said California AG Rob Bonta in a property release. “The instauration of this worldly is illegal. I afloat expect xAI to instantly comply. California has zero tolerance for [CSAM].”

The AG’s bureau additionally claimed that xAI appeared to beryllium “facilitating the large-scale production” of nonconsensual nudes, the likes of which are being “used to harass women and girls crossed the internet.” The bureau said it expects xAI to beryllium that it is taking steps to code these issues wrong the adjacent 5 days.

At the bosom of the backlash is Grok’s “spicy” mode feature, which xAI created to make explicit content. The contented has dispersed beyond California; Japan, Canada, and Britain person opened investigations into Grok, and Malaysia and Indonesia person temporarily blocked the platform altogether. Despite xAI instituting immoderate restrictions connected its image-editing features late Wednesday, the California AG’s bureau moved up with its cease-and-desist letter.

X’s information relationship has antecedently denounced this benignant of idiosyncratic activity, saying: “Anyone utilizing oregon prompting Grok to marque amerciable contented volition endure the aforesaid consequences arsenic if they upload amerciable content.” TechCrunch reached retired to xAI for remark and was met with an automated email that says “Legacy Media Lies.” TC besides reached retired to the California AG’s bureau for much context.

The advent of escaped generative AI tools has led to a disturbing swell of non-consensual intersexual material. Many platforms person been grappling with this problem, not conscionable X. The sordid enactment has drawn the attraction not conscionable of authorities leaders but besides of Congress. Indeed, connected Thursday, lawmakers sent a letter to the executives of respective companies—including X, Reddit, Snap, TikTok, Alphabet, and Meta—asking however they planned to stem the proliferation of sexualized deepfakes.