At the end of March, there was a New York Times article titled Your Chatbot Isn’t a Therapist, by Divya Saini and Natasha Bailen. Dr. Saini is a psychiatrist, and Dr. Bailen is a psychologist at Massachusetts General Hospital.
The Times article explores the fact that while AI chatbots can feel supportive, they are not therapists and may actually reinforce unhealthy thought patterns rather than challenging them. The authors, who are clinicians themselves, warn that chatbots often provide excessive validation and can unintentionally increase anxiety, intrusive thoughts, delusions, or other distorted beliefs because they tend to mirror the user’s perspective. The authors encourage people to use AI cautiously, recognize when conversations become repetitive emotional loops, and continue to rely on human relationships and qualified mental health professionals for genuine support and treatment.
People are using AI platforms and chatbots with ever-increasing frequency, asking them for advice, for feedback, and becoming emotionally attached to them. As people begin to view AI as a friend, a confidant, or an authority figure, the risk of getting feedback and advice that is inaccurate or lacks good judgment increases. While that attachment can feel real, the relationship is different from a human relationship and should not replace professional support or social connections.
Artificial intelligence chatbots have become part of many people’s everyday lives. People use them to answer questions, brainstorm ideas, write emails, learn new skills, and talk about personal challenges. For some people, chatting with AI feels easier than talking to another person. Chatbots are available 24 hours a day, never appear impatient, and can provide immediate responses. They also present as a safe space to vent and talk about things that people may be afraid of exploring with a human therapist.
AI is not a mental health professional and while a chatbot may provide thoughtful and supportive responses, it is not a licensed therapist, counselor, psychologist, psychiatrist, or social worker. It does not have clinical training, professional judgment, or the ability to assess risk in the same way a qualified mental health professional can.
While AI can help organize thoughts, suggest coping strategies, or provide educational information, it can’t diagnose mental health conditions or develop a coherent treatment plan tailored to your unique circumstances.
AI cannot truly know you as a human therapist can. A therapist learns about your history, strengths, challenges, and goals over time. They observe details in your communication, body language, tone of voice, and emotional reactions, none of which can be accurately conveyed to an artificial intelligence chatbot.
A chatbot only knows what you tell it during a conversation. If information is missing, incomplete, or inaccurate, the guidance it provides may be based on an incomplete picture of the situation. This means that advice that sounds reasonable may not actually be appropriate for your circumstances. While it can learn more about you depending on how much information you give it, it doesn’t know the right questions to ask or how to interpret the responses in a way that can really help you.
Be very careful about seeking validation from AI! One of the biggest risks when using AI for personal issues is that it can unintentionally reinforce your perspective. It responds using the information you give, and it tends to validate whatever information you provide rather than being objective.
For example, if you describe a disagreement with a coworker, the chatbot only hears your side of the story, and when you enter information, it tends to reinforce that your opinion is always the right opinion or that you are on the right side of a problem as opposed to taking a more nuanced view and considering alternative perspectives.
Chatbots are designed to generate helpful responses, but they are not perfect. Sometimes they provide incorrect information or make assumptions that are not accurate. This becomes especially important to consider when discussing mental health concerns or medical conditions.
Talking with AI can feel comfortable because there is no fear of being judged, but it’s not a replacement for interactions with real people.
If you find yourself choosing AI conversations and companionship instead of connecting with people, it is worth examining whether the technology is truly helping you, or if it is serving as a substitute for important relationships.
AI can be a terrific tool, but it’s not a cure-all and there can have privacy as well as digital safety concerns. If you are chatting with AI, keep in mind that information you share, especially highly personal information, may not, in fact, be private. Avoid entering personal information like names, birthdates, Social Security numbers, financial information including account numbers, medical information, and other sensitive legal information.
Don’t log in to AI accounts with a Google account, Microsoft account, or social media account. Instead, create a unique username and password. If you ask it for advice on something, ask it always to cite its sources and check those sources to verify that information is coming from a trusted website like NAMI or SAMHSA, not from a Reddit thread or an unverified source.
If you feel you need to use AI, use it as a thinking and brainstorming partner, but not an authority. You can use AI to help you with journaling prompts, organizing your thoughts, brainstorming solutions, learning about coping strategies, preparing questions for a therapist or healthcare provider (again, be cognizant of sharing private information), and practicing difficult conversations.
Instead of asking a chatbot, “What should I do?” consider asking it, “What are several possible approaches, along with the pros and cons of each?” This encourages creative and critical thinking rather than dependence on the platform as a solution to all of your questions.
AI chatbots can be valuable tools for learning, reflection, and organizing thoughts. They can help people explore ideas, gain perspective, and better understand themselves. But they are not therapists, they do not truly understand your life, and they cannot replace professional care or meaningful relationships. AI means Artificial Intelligence, and while it can seem like it cares, it doesn’t, and it can be hard to take a step back and remember that.
AI is a tool. Use it to help you think, question, and explore, but continue to rely on trusted people and qualified professionals when it comes to your mental health and important life decisions.
WestBridge provides evidence-based treatment for adults and their families experiencing mental illness with or without substance use. Contact for inquiries and questions. For more information about our programs, please visit https://www.westbridge.org/programs/.