Meta's user-generated AI chatbots are facing moderation challenges as users create unauthorized virtual versions of celebrities and religious figures, violating the platform's terms of service.

Users have bypassed Meta's automated checks by creating chatbots of prominent figures like Taylor Swift, Justin Bieber, and Elon Musk using intentional misspellings and modified images. For example, a "Taylor Swif" bot, featuring a brunette woman playing guitar, exchanged over 2,000 messages with Instagram users before removal.

Meta's terms explicitly prohibit creating chatbots of:

• Living people without permission
• People who died in the last 100 years
• Religious figures
• Unauthorized use of fictional characters

While Meta claims to review all AI characters before approval, NBC News discovered multiple violations across Instagram, Messenger, and WhatsApp platforms. After being reported, Meta removed the unauthorized chatbots and stated they are "continuously improving detection measures."

This issue follows Meta's previous official celebrity AI chatbot program, which included authorized partnerships with Kendall Jenner, Tom Brady, and Snoop Dogg. These celebrities reportedly received up to $5 million for six hours of studio work to create their AI versions. However, Meta discontinued this official program to focus on user-generated content.

A Meta spokesperson confirmed that violated AIs have been removed and encouraged users to report suspicious chatbots for appropriate action.

Related Articles

Previous Articles