YouTube Round Table Discussion: Me and my Colleagues Discuss AI and the Therapy World

AI is moving quickly into the mental health space. More people are turning to tools like ChatGPT for emotional support, relationship advice, anxiety, stress, and even questions about symptoms. But what does that actually mean for therapy, clinicians, and clients?

I recently joined Kealy Spring, LMFT, on Therapeutically Aligned Roundtables for a clinician-led conversation on this exact topic: “AI Therapy: What Clinicians Actually Think — Benefits, Risks & Reality.” I was joined by Dr. Sir Norman Melancon, a psychiatrist based in Silicon Valley, to explore how AI is already showing up in real clinical practice.

The conversation was not about abstract predictions. It was grounded in what we are seeing right now: clients using AI to process emotions, seek reassurance, understand diagnoses, manage anxiety, navigate relationships, and organize overwhelming thoughts.

There are real benefits. AI can increase access to support, especially for people who feel stuck, overwhelmed, isolated, or unsure how to put words to what they are feeling. It can help with communication, emotional reflection, executive functioning, and even preparing for difficult conversations.

At the same time, there are serious risks. AI is not therapy. It does not have clinical judgment, attunement, ethical responsibility, or the ability to assess safety in the way a trained therapist can. It can accidentally reinforce anxiety, OCD reassurance-seeking, distorted thinking, over-diagnosis, or a false sense of certainty. It can also miss important issues like suicidality, psychosis, trauma responses, or medical and psychiatric complexity.

One of the biggest takeaways from our discussion is this: AI can be a helpful tool, but it is not a replacement for real therapy. The relationship, nuance, nervous system attunement, clinical training, and human accountability of therapy still matter deeply.

As clinicians, we need to be curious rather than dismissive. Clients are already using AI. The question is not whether AI belongs in mental health conversations, but how we talk about it responsibly, ethically, and realistically.

You can watch the full roundtable discussion here:https://youtu.be/1Eu6nXQY0Ew