Retired psychotherapist here.
Ai therapy should be a no go area for everyone and should never be used for any personal private or confidential issues.
There are many many things that AI will never ever be able to infer or understand in the realm of human to human interaction.
Without confidentiality there can never be any trust. Trust is the fundamental foundation of any psychotherapeutic work. Trust can take months to develop, it depends on the client's past history.
You can trust your real human therapist with confidentiality because we all work within the guidelines and ethical frameworks of our professional bodies, UKCP, BACP, BPS ect.
you can complain to those professional bodies about any therapist but you cannot complain against a computer generated AI therapist. What professional bodies are ai therapists members of?
Can you imagine what will happen when six months of 50 minute therapy sessions are released online after a hack. The whole history of your personal life exposed for all to read online, things you would not even tell your husband/wife/family/children and most trusted friends.
These are the stories I have kept safe for over 20 years and would never divulge to any living human being. Could you really trust AI to do the same?
What type of professional supervision does the AI therapist use??
All therapists have to be in supervision while they work with clients. Here in the UK its one hour of supervision for every five hours of client work. The most fundamental reason for supervison is to protect the client from the therapist.
No Supervision, No client protection.
AI will never be able to observe human behaviour like another human being, the micro signs that we all understand and recognise in others, the tone of voice changes, the involuntary body movements, the micro changes to facial expressions, the way our eyes fill with tears when someone tells us a sad story, cannot be interpreted correctly by AI.
