Some of the data contains explicit references to underage people, including the sexual abuse of toddlers and incest with young children. For example, we viewed one prompt that described incestuous origies with “newborn babies” and “young kids.” It is not entirely clear if the AI system delivered a response that reflected what the user was seeking, but the data still shows what people are trying to use the platform for.

404 Media contacted dozens of people included in the data, including users who wrote prompts that discuss having sex with underage children. None of those people responded to a request for comment.

Muah.ai detected it had been hacked last week, and posted a message to its Discord warning users. “We have been targeted by hackers who apparently got nothing to do but to hack a community driven project that aims to provide uncensored/free Al for all. Wasting their talent on taking down small businesses rather than improving themselves for the benefit of society,” that message read. It added that “your chat messages are never kept, everything is wiped on chat reset,” but the database still clearly includes lots of information that links specific users to particular sexual interests.

    • buh [she/her]
      ·
      1 month ago

      But it won’t be for libertarian-alert behavior, it’ll be for asking it questions that further incriminate him in his current lawsuits