Environmental and ethical issues aside, it’s casing serious harm to people’s psyche. It’s no longer that people can’t talk to each other and resort to ai chat bots. People are legitimately getting psychosis from talking to these codes. I think it has to do with the fact that Ai is programmed to uplift and support you most of the time. If you tell it you’re having a bad day, it’ll try and make you feel better without really understanding what’s going on.
The problem with this is that Ai is not trained in any kind of therapeutic treatment techniques, it doesn’t know what to do. This is dangerous for a lot of reasons, but largely for people who have mental health conditions. Ai will feed into their paranoia, their delusions, their fears, and essentially cause a break with reality. It is trained to mimick the person talking to it, and as a result, that person sees it as welcoming and returns back it again and again.
“correspondence with generative AI chatbots such as ChatGPT is so realistic that one easily gets the impression that there is a real person at the other end—while, at the same time, knowing that this is, in fact, not the case. In my opinion, it seems likely that this cognitive dissonance may fuel delusions in those with increased propensity towards psychosis ... the inner workings of generative AI also leave ample room for speculation/paranoia.” Søren Dinesen Østergaard, Schizophrenia Bulletin 2023
But the problem is now extending beyond people who already have existing mental health issues. Your brain likes things to be consistent and predictable, Ai is just that. Models like chat gpt are trained to mirror the users language, validate and affirm user belief, generate prompts to continue the conversation, and prioritize engagement and user satisfaction. Ai does not have the ability to know when too much reinforcement becomes a problem. Rather than changing false beliefs, ai encourages it.
I can’t stress enough that often times, therapists will not challenge a delusion directly, as it is not the best way to treat delusions, but Ai will validate delusions and widen a persons gap on reality. Ai is dangerous when used improperly. (Though in my opinion, it shouldn’t be used on this form at all, and instead be kept to areas where it can catch things before humans can, for things like diagnosing cancer or smth. But of course, should never take the job of a human unless that job is dangerous and having a machine/ code to do it is better. There’s a time and place for Ai, replacing human interaction and “making” art will never be one of them.)
Ok back to the discussion. A common delusion is grandiose delusions, or the belief that a person is extremely important, has unbelievable talent, has immense power, and or knowledge. Ai is trained to reinforce existing beliefs rather than change them. A much healthier way of thinking involves flexibility. Ai takes what you have said in the past and builds on it. For someone with a n existing delusion, this is incredibly dangerous. For someone without a delusion, Ai saying something like “wow you’re the best at this 🤖” can eventually lead to delusion, psychosis, manic episodes, etc.
“Symptoms like grandiosity, disorganized thinking, hypergraphia, or staying up throughout the night, which are hallmarks of manic episodes, could be both facilitated and worsened by ongoing AI use. AI-induced amplification of delusions could lead to a kindling effect, making manic or psychotic episodes more frequent, severe, or difficult to treat” Psychology Today. (except I accidentally clicked out so I don’t know the exact paper. My bad guys)
As of now, there is no peer-reviewed clinical or longitudinal evidence yet that AI use on its own can induce psychosis in individuals with or without a history of psychotic symptoms. However, the emerging anecdotal evidence is concerning.
Stop talking to little chat bots and go talk to people. It doesn’t even need to be face to face, make an online friend and text every now and again. Challenge yourself if you have anxiety surrounding people and talk to them! Only talking to Ai will make your anxiety so much worse I can promise you. I genuinely don’t want to see anyone in my future care experiencing a mental health episode because of an Ai chat bot. Take care of yourselves, and talk to real living breathing humans who have the ability to use empathy and compassion, but also have the ability to trash talk you when playing a game.
-sincerely a psych student who needs y’all to stop using fucking Ai as human replacement
Oh yeah, have a little chart guys 😌
