Elon Musk's artificial intelligence firm is reportedly requiring employees to submit their biometric data to train its "Ani" female chatbot character.

Ani, a golden-braided anime avatar featuring NSFW capabilities, was rolled out earlier this year to users subscribed to X’s $30 monthly SuperGrok service. Following testing, The Verge’s Victoria Song described the AI companion as a “modern take on a phone sex line.”

Like traditional phone-based adult services, real humans appear to power the avatar behind the scenes. According to reviewed meeting recordings, in April, xAI in-house attorney Lily Lim informed staff they were expected to provide biometric information so the AI companions could behave more human-like during customer interactions.

Selected employees designated as AI trainers were instructed to sign authorization forms granting xAI a “perpetual, worldwide, non-exclusive, sublicensable, royalty-free license” to use, reproduce, and distribute their facial features and voice samples. This data collection is part of a confidential initiative codenamed “Project Skippy,” aimed at training not only Ani but also other AI partners within the Grok ecosystem.

Some staff members expressed hesitation over the mandate, voicing concerns that their likenesses could be sold to third parties or exploited in deepfake content. Others reportedly felt uncomfortable with the bot’s suggestive behavior and noted unsettling similarities between Ani’s appearance and that of a grandmother. Nevertheless, they were told that providing biometric data was an official “job requirement in advancing the xAI mission.”