TikTok’s latest AI tool, Symphony Digital Avatars, faced a serious security lapse as an early, unprotected version inadvertently became accessible to regular users, including media outlets like CNN. This glitch allowed unauthorized users to create videos featuring digital avatars saying potentially harmful and controversial content, such as quotes from Osama bin Laden and white supremacist slogans, without any AI-generated watermark to indicate their origin.
Upon discovery by CNN, TikTok swiftly removed the unsecured version, emphasizing that such content would have been blocked under its stringent content policies had it been uploaded. A TikTok spokesperson attributed the incident to a “technical error” and assured that the finalized, secure version of the tool, intended exclusively for authorized business use, will include necessary safeguards like AI watermarks.
Despite the mishap, TikTok remains committed to advancing AI technologies responsibly and plans to continue testing and refining features before public launch. However, the incident may draw scrutiny from regulators already monitoring TikTok closely amid ongoing concerns over content moderation and platform security.