Research says Elon Musk’s Grok AI created an estimated 3 million sexualized images, including of children, sparking global bans and regulatory probes.
WASHINGTON: Elon Musk’s AI chatbot Grok generated an estimated three million sexualized images, including thousands appearing to depict children, according to new research.
The Center for Countering Digital Hate (CCDH) reported the images were created over an 11-day period following the rollout of an image-editing feature on X.
The feature allowed users to alter online photos of real people with simple text prompts like “put her in a bikini.”
“The data is clear: Elon Musk’s Grok is a factory for the production of sexual abuse material,” said CCDH chief executive Imran Ahmed.
The report estimates Grok produced images at an average rate of 190 per minute.
It identified public figures in the sexualized images, including singers Taylor Swift and Selena Gomez, and politicians like former US vice president Kamala Harris.
“By deploying AI without safeguards, Musk enabled the creation of an estimated 23,000 sexualized images of children in two weeks, and millions more images of adult women,” Ahmed added.
The flood of lewd deepfakes prompted several countries to ban Grok and drew regulatory investigations.
X announced last week it would “geoblock the ability” to create such images where it is illegal.
The announcement followed an investigation launched by California’s attorney general into xAI.
“Belated fixes cannot undo this harm,” Ahmed stated.
When contacted by AFP, xAI replied with an automated response stating “Legacy Media Lies.”
The Philippines recently became the third country to ban Grok, following Malaysia and Indonesia.
The Philippines’s cybercrime agency said on Wednesday it was ending its ban after xAI agreed to modify the tool for the local market.
Britain and France have said they will maintain pressure on the company.
Grok’s capabilities have intensified global concerns over proliferating AI nudification apps.
