AI deepfakes on X raise a major policy question

Sexualized deepfakes have led to some of the most trenchant criticism of generative artificial intelligence (AI) after affecting everyone from Taylor Swift to high school students. Grok, the AI chatbot and image generator launched by Elon Musk, has effectively turned this disturbing phenomenon into a meme over the past week. Trollish provocateurs kicked off the trend after a December 2025 update made it easier to use Grok to remove clothing from images. Users have directed the chatbot to digitally undress people in innumerable images, which Grok then posts to X. Those affected have primarily been women, and there have also been cases of sexualized AI images of children being distributed on the platform. Governments around the world are scrutinizing X, from France to Malaysia to India. But should it be X that’s on the hook, or its users? Musk, who owns X, has suggested that the users themselves should face legal consequences, and online speech advocates have taken a similar position. Advocates for victims of sexual exploitation argue that platforms should be held accountable, too.