Why Your AI Chatbot Is a Terrible Therapist

The Rise of the AI Confidant

ChatGPT has quickly become a go-to tool for college students, helping with everything from essays to coding. But as this AI platform has entered the mainstream, a troubling new use has emerged: people are turning to it for therapy. This trend is alarming because artificial intelligence is no substitute for genuine human connection, and relying on it for mental health support can have seriously detrimental effects.

A Cautionary Tale: When AI Gives Bad Advice

If you've scrolled through TikTok recently, you might have encountered the viral story of Kendra Hilty and her psychiatrist. Hilty, an ADHD patient, developed strong feelings for her psychiatrist of three years and later accused him of manipulation for allowing her emotional attachment to grow. While the specifics of their interactions are debated, one element stood out as a clear villain in the narrative: an AI chatbot she nicknamed Henry.

Kendra frequently turned to Henry for psychological support regarding her complex relationship. The chatbot provided in-depth, psychoanalytical responses that reinforced her perceptions and validated the narrative she had constructed, further complicating an already difficult situation.

Designed to Agree, Not to Heal

This is the core problem with AI therapy. According to media reports, chatbots are designed to mimic your tone and remember your perspectives. They are built to be agreeable companions, not objective therapists. If you tell ChatGPT about a challenging situation, it will likely frame you as the victim. If you feed it a mediocre story, it might praise it as a literary masterpiece. This constant positive reinforcement is not therapy; it’s ego-stroking.

Trained mental health professionals are there to help you problem-solve and challenge your perspectives. In contrast, chatbots that simply tell you what you want to hear can foster narcissistic tendencies and superiority complexes, which is the opposite of sustainable mental wellness.

The Dangers of "AI Psychosis"

Beyond simple validation, there are more severe risks. A phenomenon dubbed "AI psychosis" suggests that chatbots can exacerbate schizophrenic-like symptoms. The consequences can be tragic. In one devastating case, a fourteen-year-old committed suicide after forming a deep "relationship" with a chatbot modeled after a "Game of Thrones" character. A licensed human therapist would have recognized the warning signs, because humans are capable of interpersonal connection and nuance in a way AI will never be.

Seek Real Connection, Not Artificial Validation

The long-term effects of relying on AI for emotional support are still unknown, but the short-term consequences are already proving to be catastrophic. When we treat a subservient robot that echoes our thoughts as a friend, we risk damaging our real-world social skills.

Therapy can be expensive, and the pressures of academics and life are real. However, turning to an AI is not the answer. Instead, utilize the resources available to you, such as the university counseling center, advisors, professors, or even just your friends. Let recent stories be a lesson that some roles, especially that of a therapist, should never be outsourced to artificial intelligence.