With that context, I can't see a way in which ERP therapy with an LLM (at least, the consumer system prompts we are used to and often think of) would not backfire and be actively harmful.

There are two reasons, and both relate to the people-pleasing tuning that ChatGPT etc have:

1. ERP is actively uncomfortable, that is the point. But there is a delicate balance to the level of discomfort.

Too comfortable and you make no progress. Too uncomfortable and you can set yourself back enormously. We aren't talking one step forward and two back. I got overly confident two months in and put myself in a worse state than I had been on entering therapy (though not my worst state overall over the 20 years). I haven't seen many people in support groups do very successful self-administration of ERP for this reason. My therapist had to do a lot of non-verbal reading of my reaction to gauge if I was at an appropriate level of discomfort, as it was not something I was able to verbalise myself without a lot of practice and learning - and that learning and practice came from guidance by a human who didn't need me to be able to articulate it already.

2. Reassurance seeking is a compulsion, and it is one of the most common and difficult to stamp out.

I can't see an LLM not providing reassurance when asked for it, and so I can't see anyone using an LLM for therapy as making any progress at all without a level of discipline and awareness they would have had to obtain from in person ERP anyway.

The day just started and I'm already done with it.

- I want to do a thing.

- Do the thing.

- but I’m scared

- Do it scared.

  Luckily, LLMs significantly reduce the effort/cost of therapy experiments. Consider trying the following prompt:

  > Please guide me through a round of ERP therapy. Start by listing universal sources of fear/discomfort/anxiety.

  If you find this process useful, consider trying it with a licensed human professional.

Do you know of any low-friction ways to systematically tackle fears/discomforts? I really want to recommend a quick experiment that folks can try at home without doing full-blown therapy.

Even in "Realms of the Human Unconscious: Observations from LSD Research Paperback" by Stanislav Grof, OCD gives remarkable resistance not only to therapy, but even to psychedelics. This book discusses a case of a person with OCD who took 500ug of LSD (well, a rather high dose) yet neither the patient nor their psychiatrist noted any effect. Yet, after getting at ease, much smaller dose worked (and there was a remarkable progress).

What might work though, is reducing overall stress revel. OCD symptoms are often a response to anxiety and uncertainty.

The way I usually pitch the book is that Mark is not an MD, or a scientific expert on the topic, but he is a patient who's been through it, is very interested, and explains to you how it worked for him, and what he knows about the topic like a friend would. There are some mistakes, some things are a bit awkwardly explain, but the book is overall a tremendous read for anyone interested in anything dealing with anxiety, ocd, or more generally a mind too keen on spinning in a wheel.

https://www.markfreeman.ca/books/

Example:

You are afraid of dogs. LLM tells you to approach one to habituate your anxiety. You do that but the fear overwhelms you and you run away.

Brain’s diary: I avoided death, that beast would shred me to pieces in a moment.

(Note this is about „normal” anxiety and not OCD, but the idea translates.)

If you dont want (or e.g. cannot afford) a proper therapy, IMHO a book by a competent author is a much better choice than LLM. Or a few different books.

It is literally “My Mental Health is Improving Through Human Contact: Human Contact Might Not Be For You. The Only Way To Know Is By Asking The Computer”

It is kind of like “I started exercising with a trainer and I feel better and I’m losing weight, because of this I suggest you take up smoking and based on that decide whether or not you want to exercise”. Like the advice is simultaneously almost completely unrelated to their experience and also the exact opposite of their experience.

I cannot wait to hear more well-thought-out advice from the author of “God created men; Sam Altman made them equal”

This set me off on a harder path of more abstract therapy (I'm working with an integrative therapist who practices across IFS, Jungian etc), meditation etc. The book the untethered soul, by Michael Singer helped a lot to put everything into perspective - the therapy, the anxiety, the day to day - in a way that nothing else had really achieved for me.

Also, a key conclusion of Cartwrights is that individualised change/treatment is important, but it's worthless without community. I think she's deeply, profoundly right.

CBT/ERP, medicalised Skinner pavlovian model, cheap, short term relief of symptoms, no long lasting effect.

CBT = there is no unconscious!

CBT is based on a theory that it is not events in themselves that upset us, but the meanings we give them. However, CBT believes that this meaning is conscious, can be accessed, and is not ambiguous.

If someone says that they feel sorry for a person, they do feel sorry and it is possible that their problem is that they should not; the therapist then tries to help the patient to see that they are thinking about something in the ‘wrong way’. By using brain washing techniques.

The concept of the unconscious, a central Freudian discovery, renders the picture instantly more complex: if you say you are sorry for someone it might indeed mean that you are sorry, but could also mean that you can’t face your own aggressiveness and the fact that you are delighted at what happened to the other.

The question of the existence of the unconscious is crucial, because if CBT starts to take into account the possibility of unconscious meanings and logic, then it would become another branch of psychoanalysis.