AIHealthcare Analytics

Overview

AI applications in mental health — including chatbot-based therapy, symptom triage, and crisis support — face mounting scrutiny regarding clinical efficacy, safety, and governance in 2026.

Columbia University: Chatbots as Therapists

Columbia University published an analysis titled "Your Chatbot is a Terrible Therapist. Here's Why." (article 14), providing academic critique of AI chatbot deployment in therapeutic contexts. The analysis identifies structural limitations of current LLM-based chatbots that prevent them from functioning effectively as therapists — including lack of longitudinal context, inability to detect non-verbal cues, and absence of clinical accountability.

Implications for Governance

The Columbia critique reinforces the case for mandatory clinician oversight in AI mental health applications (see ai-mental-health-clinician-oversight) and the broader governance challenges identified in state-level legislation and regulatory frameworks. The proliferation of consumer mental health AI tools without clinical oversight remains a significant policy gap.