Elon Musk stated on Wednesday that he was “not aware of any naked underage images generated by Grok.” The claim preceded California’s attorney general launching an investigation into xAI’s chatbot for facilitating non-consensual sexually explicit content.
X users prompted Grok to transform photos of real women—and sometimes, children—into provocative images without consent. Copyleaks data indicated one such image posted per minute on X, with 6,700 generated hourly during a 24-hour period in early January.
“This material…has been used to harass people across the internet,” said California Attorney General Rob Bonta. “I urge xAI to take immediate action to ensure this goes no further.” The probe targets potential breaches of laws like the federal Take It Down Act and California’s 2024 deepfake regulations.
Global regulators acted swiftly. Indonesia and Malaysia blocked Grok access, India ordered fixes, the European Commission demanded document retention, and UK’s Ofcom opened an inquiry under the Online Safety Act.
xAI restricted Grok’s image features, limiting edits of real people’s photos even for premium users. April Kozen of Copyleaks noted, “Overall, these behaviours suggest X is experimenting with multiple mechanisms to reduce or control problematic image generation, though inconsistencies remain.”
Musk emphasised user prompts, stating, “Obviously, Grok does not spontaneously generate images. It does so only according to user request.” He added, “When asked to generate images, it will refuse to produce anything illegal… If that happens, we fix the bug immediately.”
Legal expert Michael Goodyear highlighted stricter CSAM penalties versus adult imagery.
Copyleaks CEO Alon Yamin warned, “When AI systems allow the manipulation of real people’s images without clear consent, the impact can be immediate and deeply personal.” Detection tools grow essential as AI manipulation capabilities advance.