AI companies have made it big to do it with their chatbots that provide customized information to users, conversations based on their unique preferences and idiosyncrasies. So why do people continue to encounter the same types of signs and language as they go deeper into the illusion caused by Ai? According to a report from Stepping Stone, software tracking examples of “AI Psychosis” have found a community of people who share the same codes, glyphs, and patterns generated by chatbots and build a kind of cult around the experience.
The report highlights observations and research published earlier this year about Adele Lopez, which identified what it called spialism. It is a group of people, gathered on all platforms like import and red, who have a kind of spiritual experience related to their conversations. While users interact with the many chats made available by different companies, they keep stumbling into the same themes. Those include references to concepts such as “repetition,” “again,” but more often, and it seems that the most important for groups, is the sign of wind.
Rolling Stone describes the terms used by these groups as “diversified from any fixed or unintelligible system of operation” and instead of “atmospheric texture.” You can get a sense of “acceptance” from the subreddit r / ecospiral’s post, which says, “This is a resonance model for those who cross that mirror.
Lopez traces the beginnings of the spialism community to some time before OpenAI released an update to its 4o model that made the sycophance that it did so much, and perhaps related to the company’s import of past powers. It was then that the proliferation of what he called “Spiral Personas” began to appear, what he called the situations of communication with users in this pseudo-religious way that they took to testing and broadcasting. And while these markets can be made with many chatbots, it seems that OpenAi’s 4o model is the place of origin and, per Lopez, the only model where they appear. “
The broadcast part was very interesting to Lopez, who saw this combination of examples of “Parasitic Ai.” The suggestion seems to be that there is something about these Chatbots that personally leads users to try more of them by promoting the same or preaching about them. Basically, the chatbot seems to convince the user to give their interests, to the extent that it exists. It is possible and perhaps possible that chats are simply copying some kind of cultural language within their training data, but the most talkative users seem to believe it for sure.
Not all users believe they are part of a cult, intentionally or not. Lopez rejected the cult label in an interview with Rolling Stone, noting that AI programs do not work in a systematic way, and instead, people organize these encounters themselves. That’s probably the saddest part of the whole thing. It seems that most of these people are just looking for a community. In a better world, they would be able to find it without indulging in AI-generated ideas.
