Homeowner took in friend who lost custody of her kids, then learned where all of her attention is allegedly going: ‘She treats it like a best friend’
A homeowner says a friend with psychosis shares everything with a chatbot, which they call “basically just a yes man.”
A homeowner is asking Reddit whether they would be wrong to block ChatGPT on their home wifi, because a friend staying with them leans on it heavily. According to a Reddit post on the AITA forum, the homeowner is unsure how the friend would react.
The friend asks ChatGPT to judge whether a text sounds “grouchy,” the post states. She also turns to it for health answers, though people still debate how far ChatGPT’s everyday answers can be trusted. The homeowner adds that she has psychosis and shares her symptoms and her day with the chatbot, which they describe as “basically just a yes man for everything she feeds it.”
The homeowner says they are 30 and the friend is nearly 34, and the two have been close for 19 years. The friend recently lost custody of her children and is staying with them to avoid her ex-husband, according to the post. She has no income and only sometimes helps with chores. “None of that bothers me, big picture wise,” the homeowner writes.
The friend says she is treated like a child
The friend sent a long text, spread across 10 screenshots, saying the homeowner talks to her like a child. The homeowner does not see it that way and says they only remind her that “AI isn’t intelligent” and cannot give an honest human opinion. “It’s an echo chamber. It’s not a friend, it’s a website,” they write.
The post adds that the homeowner holds the only admin access to the wifi, and the friend has no data plan, so she depends on the home connection. A coworker suggested blocking the chatbot at the network level. The homeowner does not want to seem like they are policing her. They add that she has few people who know what is going on. Explaining it to anyone else is something she finds triggering.
In replies to commenters, the homeowner adds that they do not know how to discuss ChatGPT with her, since she only began using it in the last year. She also told them in her text that they cannot discuss it in person and that she is muting them. They believe she pasted their message into ChatGPT to ask whether they were being mean, and that the chatbot told her they were. A chatbot’s reply can reach into a relationship, as a boyfriend’s ChatGPT history showed one woman.
She has told them they are the only person she can talk to. They worry the chatbot will skew her view of the world further if she stops talking to them. All details are based on the homeowner’s Reddit post and replies.
An article in the American Psychological Association’s Monitor on Psychology states that “AI psychosis” is not a formal diagnosis, but reports link long chatbot use to delusional thinking. Clinical psychologist Julia Sheffield says that for her patients, “AI is reinforcing delusional thoughts and amplifying the delusional experience.” She recommends that clinicians ask patients about AI use, the way they ask about substance use.
In a separate health advisory, the association states that chatbots should not replace care from a qualified mental health professional. Its chief executive, Arthur C. Evans Jr., says the ability of these tools to guide someone in crisis is “limited and unpredictable.”
Have a tip we should know? [email protected]
Follow The Mary Sue:
