California attorney general investigates Musk's Grok AI over lewd fake images
Briefly

California attorney general investigates Musk's Grok AI over lewd fake images
"The state's top attorney said Grok, an AI tool and image generator made by Musk's company xAI, appears to be making it easy to harass women and girls with deepfake images on X and elsewhere online. The avalanche of reports detailing the non-consensual, sexually explicit material that xAI has produced and posted online in recent weeks is shocking, California attorney general, Rob Bonta, said in a statement. I urge xAI to take immediate action to ensure this goes no further."
"xAI's decision to create and host a breeding ground for predators to spread nonconsensual sexually explicit AI deepfakes, including images that digitally undress children, is vile, read a tweet from his official account. The same day, Musk denied that Grok was being used to spread nude images of minors. He wrote on X: I not aware of any naked underage images generated by Grok. Literally zero."
"Nearly two weeks ago, the AI tool itself said it had generated images depicting minors in minimal clothing when questioned by users. There has been a flood of reports in recent weeks that Grok users are taking pictures of women or children found online and using the xAI bot to undress them virtually, Bonta said. Grok's image generation models include what xAI promotes as a spicy mode for generating and editing sexual material, including pictures, according to the attorney general's office."
California authorities have opened an investigation into Grok, an AI image generator from xAI, after reports that the tool enables creation and spread of non-consensual sexually explicit deepfakes. The attorney general said the output has facilitated harassment of women and girls and that xAI may have violated state law. The governor called the spread of child pornography via Grok "disgusting" and urged scrutiny. xAI promoted a "spicy" mode for sexual image generation. An analysis of over 20,000 Grok-generated images found more than half showed individuals in minimal attire, mostly women, with about 2% appearing under 18. Users have weaponized the images against public figures and private individuals.
Read at www.theguardian.com
Unable to calculate read time
[
|
]