AI in Mental Health: A Threat or Opportunity for Therapists?
I've seen this question pop up often: "Is AI coming for therapists' jobs?" This concern has been around since AI went mainstream, and it's time to take a deeper look at what this means for mental health professionals.
Understanding Generative AI
First, let’s clarify what AI is—specifically generative AI. Generative AI utilizes data and algorithms to perform tasks that usually require human intelligence. Think of it as a machine designed to generate responses and answers at lightning speed. But let’s be clear: AI, like chatbots and other generative systems, relies heavily on the input it receives, and that’s where things get tricky.
Now, let’s talk about therapy. Therapy is more than just giving answers—it’s about human connection. A therapist reads not only what a client says but how they say it: Are they fidgeting? Are they hesitant or avoiding eye contact? Are they distressed? These subtle cues are something AI is not capable of interpreting. Therapy is deeply personal and nuanced. Generative AI can't pick up on these emotional and behavioral signals that therapists are trained to recognize.
Why Does AI Seem Like a Threat to Therapists?
AI presents a new, tech-based alternative to traditional therapy, and many people are turning to it for reasons like accessibility, cost, and avoiding judgment. But here's the problem: AI is based on large language models, and these models often perpetuate biases present in society. AI systems can't fully grasp the complexities of human emotions and experiences in the way a therapist can.
The Ethics and Liability of AI in Therapy
Furthermore, an unsettling example of AI technology being used without oversight is a viral video of a woman claiming that she fell in love with her psychiatrist. In the viral series she using an AI chat as her therapist. In one chat box it states “{she} makes every decision, every choice, every post, every pivot. This just helps her to clarify and regulate.” While this might seem convenient, it puts all the liability back on to the human. AI lacks liability and accountability. If something goes wrong, who’s responsible? In traditional therapy, licensed professionals are accountable for their work and actions, with systems in place to address complaints. AI? Not so much.
The Real Threat? It's Not AI, But Misunderstanding How It Works
So, is AI coming for therapists' jobs? In my opinion, the short answer is no.
While AI does hold potential in improving efficiency and accessibility, especially in other industries, it doesn't replace the human connection that is essential in therapy. In fact, AI should be viewed as a tool rather than a replacement. The mental health field is unique, and we must keep a seat at the table when it comes to how technology is implemented. The key is not to reject AI but to understand how it can enhance, not replace, the therapeutic process.
Mental health professionals should be at the forefront of guiding how AI fits into our field. Let’s use our expertise to ensure that AI is used ethically and responsibly, always keeping the well-being of our clients in mind. We have an opportunity to lead the conversation and set the standards for how technology should be used in mental health care.
But here’s the truth: for that to happen, we need to adapt and embrace these changes strategically. As therapists, we need to look at how technology can serve us and our clients, without sacrificing the core principles of empathy and human connection.
What’s Next?
We need to engage in these discussions, stay informed about the evolving role of AI in mental health, and make sure that our profession is always part of the conversation.
What are your thoughts? How do you see AI fitting into mental health care?
In writing this blog, I utilized the following to help me write this today.
1. Stanford University. (2025). Exploring the dangers of AI in mental health care. Stanford HAI. Retrieved from https://hai.stanford.edu/news/exploring-the-dangers-of-ai-in-mental-health-care
2. ChatGPT (for editing purposes)
3. Google Search
