That is one bullshit headline. Forbes keeping the AI pump and dump scheme going.
TLDR: People correctly discerned that written responses were from an “AI” chatbot slightly less often than they correctly discerned that responses were from a psychotherapist.
“AI” cannot replace a therapist and hasn’t “won” squat.
Holy shit, thanks for saving us the click. Wtf
A bit disingenuous not to mention this part:
Further, participants in most cases preferred ChatGPT’s take on the matter at hand. That was based on five factors: whether the response understood the speaker, showed empathy, was appropriate for the therapy setting, was relevant for various cultural backgrounds, and was something a good therapist would say.
Patients explaining they liked what they heared - not if it is correct or relevant to the cause. There is not even a pipeline for escalation, because AIs don’t think.
Exactly. AI chatbot’s also cannot empathize since they have no self awareness.
You can’t say “Exactly” when you tl;dr’d and removed one of the most important parts of the article.
Your human summary was literally worse than AI 🤦
I’m getting downvoted, which makes me suspect people think I’m cheerleading for AI. I’m not. I’m sure it sucks compared to a therapist. I’m just saying that the tl;dr also sucked.
but it can give the illusion of empathy, which is far more important.
You’re doing the lord’s work 🫡
I don’t trust therapists, but I trust OpenAI with my mental state about 100x less.
Did they compete on providing actual therapy? No? Then this is meaningless.
Imitating a therapist is not too hard, look up Eliza. Was it good as an actual therapist? Haha, no.
Or Doctor Sbaitso.
I’ve used AI as a pseudo-therapist. It was kinda surreal. It had some helpful things to say, but there was a whole lot of cheerleading. Like, I appreciate the boost, and telling me how great I am. Then it kept trying to push me into an action plan like it’s selling a Tony Robbin’s book. And it never really challenged me on my representations or perspective except when I was down on myself.
I get it, when someone comes to you with troubles, try to make them feel better about themselves. But I really have to do a lot of searching to figure out what parts are worth paying attention to and what parts are just hyping me up.
I definitely would not trust it, but I think it says some useful stuff by accident now and again.
Maybe it would’ve done better if I’d given it really detailed instructions on how to be a therapist, but if I could do that I could probably give those same instructions to my wife or someone and be better off.
From the study
Using different measures, we then confirmed that responses written by ChatGPT were rated higher than the therapist’s responses suggesting these differences may be explained by part-of-speech and response sentiment. This may be an early indication that ChatGPT has the potential to improve psychotherapeutic processes. We anticipate that this work may lead to the development of different methods of testing and creating psychotherapeutic interventions. Further, we discuss limitations (including the lack of the therapeutic context), and how continued research in this area may lead to improved efficacy of psychotherapeutic interventions allowing such interventions to be placed in the hands of individuals who need them the most.
