TLDR: Mental health therapists are increasingly incorporating generative AI tools, including large language models, directly into client therapy sessions. This emerging practice, while offering potential benefits for enhancing therapeutic effectiveness, is considered controversial as it challenges traditional client expectations of human-only interaction and raises questions about transparency and ethical implications.
A new trend is emerging in the field of mental health, where therapists are reportedly integrating generative Artificial Intelligence (AI) and large language models (LLMs) directly into their client therapy sessions. This development, highlighted in a recent Forbes column by Dr. Lance B. Eliot, a renowned AI scientist and consultant, is stirring considerable professional debate due to its controversial nature.
Traditionally, clients expect a purely human interaction during therapy. The notion of a therapist leveraging AI during a session is a significant departure from this norm. While AI tools have been used by therapists for preparatory work before sessions or for organizing notes and seeking ‘second opinions’ afterward, the active employment of AI during an ongoing session presents a new frontier.
According to Eliot’s analysis, therapists are utilizing various LLMs, ranging from popular generic platforms like ChatGPT, GPT-5, Claude, Grok, Gemini, and Llama, to specialized or customized LLMs specifically designed for mental health applications. The integration can be subtle, with AI potentially listening to the therapeutic dialogue and providing real-time, instant commentary or reminders to the therapist, often without the client’s explicit awareness.
Proponents of in-session AI use argue that it can significantly enhance a human therapist’s performance. The AI can serve as an invaluable assistant, prompting the therapist with relevant information or insights, thereby improving the quality of care. As Eliot suggests, if AI were confined to pre- or post-session use, its real-time potential to ‘make the magic happen’ during the actual interaction would be lost. The presence of both therapist and client, augmented by AI, is seen by some as the optimal moment for AI to offer its contributions.
However, this practice is not without its complexities and risks. Dr. Eliot, who has extensively covered the advent of modern-era AI in mental health, acknowledges the tremendous upsides but also warns of ‘hidden risks and outright gotchas.’ The ethical implications, particularly regarding client consent, data privacy, and the potential impact on the human-centric nature of therapy, are central to the ongoing discussion. The expectation of an undivided, empathetic human presence from a therapist is deeply ingrained, and the introduction of AI into this intimate space challenges those foundational assumptions.
Also Read:
- Artificial Intelligence Poised to Revolutionize Health Economics and Care Delivery, New Research Indicates
- Maryland Grapples with AI ‘Hallucinations’ in Legal Cases, Spurs Legislative Push for Guardrails
This evolving landscape underscores a broader shift in how technology intersects with sensitive professional practices, prompting a reevaluation of therapeutic boundaries and the role of AI as a tool in mental healthcare.
