France's AI and digital ambassador Clara Chappaz says making public image generation a paid feature of Grok is a "scam", adding to the outcry over how tech mogul Elon Musk has dealt with a torrent of deepfake sexual abuse on his social media platform X.
After Malaysia and Indonesia suspended access to Grok this weekend, French foreign ministry official Chappaz declined to say whether France would follow suit, but indicated that it was working on an international response, and said she hoped the courts would mete out swift justice for what she called a "totally illegal" use of AI.
"Everyone in France who's been a victim of this should definitely take it to the courts," Chappaz said, underlining that in France, generating non-consensual sexual deepfakes is punishable by up to three years' prison and €75,000 in fines.
The Telegraph, a British newspaper, reported this weekend that Australia, the UK and Canada are considering joint action – though Canada’s AI minister Evan Solomon has since said in a post on X that the country won’t be banning the platform.
Since the start of the year, X users have flooded the platform with sexually abusive, violent, and extremist content generated by the artificial intelligence chatbot Grok, with users notably asking it to alter pictures of women in order to strip them of their clothes.
After widespread criticism from governments and civil society, X put the image generation function behind a paywall for those using it within public-facing replies and posts on X. But the tool remains available for free in a private Grok area of the website.
Since this change, researchers have indicated that the number of illicit posts has declined from its peak of tens of thousands a day in the first week of January.
Chappaz called the tweak "completely hypocritical".
"I see it as a scam because what it means is people pay to get access to the functionality, and guess who benefits from it? X, who's getting more income as a result," she said.
English (US) · 