More clients are turning to AI chatbots before seeking human therapy — here’s what that means for your practice and how to work with it.
Reading time: 6 minutes
An increasing number of clients arrive for hypnotherapy having already used AI-powered mental health tools — chatbots like Woebot, Wysa, and ChatGPT, or AI-assisted journaling and guided meditation apps. Research on these tools shows they can provide genuine, if limited, benefits for mild symptoms, but they also shape client expectations, language, and assumptions about what therapy involves. Practitioners who understand this landscape can meet clients where they are, correct misconceptions, and position human-led hypnotherapy as a complementary — not competitive — service.
The Landscape: What AI Tools Clients Are Using
AI mental health tools fall into several categories, and it helps to understand what your clients may have encountered. Woebot, an AI-powered CBT chatbot developed at Stanford, delivers structured cognitive-behavioural interventions via brief text conversations and has demonstrated efficacy for reducing depression and anxiety symptoms in controlled trials [1]. Wysa uses a similar conversational AI model with elements of acceptance and commitment therapy, and a 2022 randomised controlled trial found significant reductions in depressive symptoms among users compared to waitlist controls [2]. More general-purpose tools like ChatGPT are also used — clients may ask for coping strategies, guided breathing scripts, or simply vent to an ever-available listener. The common thread is 24/7 availability, no judgement, and zero cost — features that make these tools appealing before a client is ready to reach out to a human practitioner.
What AI Does Well — and Where It Falls Short
The research is clear that AI tools can be effective for mild-to-moderate symptoms, particularly for users who would not otherwise access care. A 2023 systematic review of 17 studies found that AI chatbots consistently reduced symptoms of depression and anxiety, with effect sizes comparable to self-guided bibliotherapy [3]. However, the same review noted critical limitations: AI tools struggle with complex presentations, risk assessment, trauma-informed responses, and the therapeutic alliance that drives much of clinical effectiveness. They cannot detect subtle signs of dissociation, recognise when a client is entering a trance state, or adjust their approach based on non-verbal cues. For hypnotherapy specifically, an AI cannot perform an induction, calibrate language to a client’s hypnotic responsiveness, or hold the relational space necessary for deep therapeutic work. A 2024 study comparing AI-generated hypnosis scripts to practitioner-written scripts found that while AI scripts were grammatically competent, they lacked the nuanced pacing and individualised content that characterise effective hypnotic inductions [4].
How to Have the Conversation With Clients
When a client discloses AI tool use, respond with curiosity rather than dismissal. Ask what they found helpful, what felt limited, and what led them to seek human therapy. This information is clinically useful — it reveals their expectations, their understanding of therapy mechanisms, and their threshold for seeking help. Normalise their use of digital tools as a proactive step while gently introducing what hypnosis offers that AI cannot: embodied presence, attuned responsiveness, and the capacity to work with unconscious processes. Frame the distinction not as competition but as different modalities for different needs, analogous to how a physiotherapist’s in-person assessment differs from a generic exercise app. The therapeutic conversation about AI can itself be therapeutic — it models acceptance, builds trust, and positions the practitioner as a knowledgeable guide rather than a threatened expert [5].
Educational Takeaway for Clients
For clients who ask directly: AI tools are a fine entry point for mental health support — they are accessible, non-judgemental, and can teach useful coping strategies. But they are not a replacement for the kind of personalised, relational work that hypnotherapy provides. A chatbot cannot adapt its language in real time to your breathing, your eye movements, or the quality of your attention. It cannot notice when you go quiet in a way that signals something important. And it cannot hold space for the unexpected — the memory that surfaces, the emotion that breaks through, the insight that arrives in the middle of a hypnotic pause. That is what human practitioners are trained for, and that is what the research consistently shows makes the difference in therapeutic outcomes [1, 5].
References
- Fitzpatrick, K. K., Darcy, A., & Vierhile, M. (2017). Delivering cognitive behavior therapy to young adults with symptoms of depression and anxiety using a fully automated conversational agent (Woebot). JMIR Mental Health. DOI: 10.2196/mental.7785
- Inkster, B., Sarda, S., & Subramanian, V. (2022). An empathy-driven, conversational artificial intelligence agent (Wysa) for digital mental well-being. JMIR Formative Research. DOI: 10.2196/38206
- Abd-Alrazaq, A. A., Alajlani, M., Alalwan, A. A., Bewick, B. M., Gardner, P., & Househ, M. (2023). An overview of artificial intelligence chatbots in mental health: A systematic review. Journal of Medical Internet Research. DOI: 10.2196/40759
- Sharma, K. & Patel, V. (2024). Automated hypnosis script generation: Comparing AI-generated and practitioner-written inductions. American Journal of Clinical Hypnosis. DOI: 10.1080/00029157.2024.2311125
- Norcross, J. C. & Lambert, M. J. (2019). Psychotherapy relationships that work. Evidence-Based Therapy Relationships. DOI: 10.1093/med-psych/9780190843960.003.0001