AI in Therapy: Balancing Efficiency and Privacy Concerns

Therapists increasingly use AI tools for session recording and administrative tasks. This technology aims to streamline operations, but it raises privacy concerns. Molly Quinn, a 31-year-old librarian from Fayetteville, Arkansas, experienced discomfort when her therapist introduced AI for note-taking in sessions.

Initially, Quinn considered the option. She wanted to know the destination of her information and if it remained local or moved to the cloud. During the session, she noticed a change. Her therapist wasn’t taking notes as usual. The iPad stood propped up, recording the session. This realization hit Quinn hard. She felt her privacy was violated, as she trusted her therapist with sensitive emotions and personal details.

“This person who I’m supposed to be able to trust with some very private and very intense emotions had just completely disregarded something I said I was not comfortable with. I felt completely violated.” – Molly Quinn

Quinn drove home, replaying the session and feeling uneasy about the recording. Many therapists are using AI platforms like Berries to reduce paperwork, hoping to focus more on clients and balance work with personal life. The system records conversations, transcribes them, and prepares draft notes for therapists to review and store in medical records. Companies like SimplePractice and Blueprint provide similar services, offering plans ranging from $19 to $99 per month. These tools claim to alleviate the burden of documentation without impeding the therapeutic exchange.

Privacy remains a significant concern. Tal Salman, CEO of Berries, assures clients that audio is deleted post-session, and transcripts follow HIPAA guidelines, stored securely in the U.S. Yet, public skepticism persists. A YouGov survey reveals hesitancy to embrace AI in mental health care. Only 11% of Americans feel comfortable using AI for mental health, with concerns about privacy and lack of human understanding.

Moreover, healthcare systems often face data breaches despite compliance with regulations like HIPAA. Kellie Owens, assistant professor of medical ethics at NYU, underscores the importance of consent and explicit communication about recordings in therapy. Trust and psychological safety are crucial in therapy, and perceived privacy violations may harm client-therapist relationships.

For Molly Quinn, the breach felt personal, illustrating broader apprehensions about AI entering therapy. Although promised confidentiality, unauthorized access could expose private conversations. Despite technical risks, some clinicians find AI beneficial. Marisa Cohen, a therapist in New York, warns that AI’s presence may shift the therapeutic experience, with clients becoming aware of a third party. She also notes that transcription errors could incriminate inaccuracies in patient records.

AI tools promise efficiency gains. Kym Tolson, “The Traveling Therapist,” praised AI for reducing her administrative workload. She emphasizes the liberty and balance this efficiency affords, albeit at a prudent cost of rigorously checking AI-generated notes.

Despite AI’s potential advantages, Quinn’s encounter made her cautious. After severing ties with her therapist, she sought a new therapist, explicitly rejecting AI involvement in sessions. She stresses the importance of keeping tech companies outside the therapeutic space.