Should I tell my therapist I use an AI companion?

Yes. Tell them, and tell them early. What you type into an AI companion at 1 a.m. is part of your mental health picture, and a therapist working without it is working from an incomplete file. Most therapists in 2026 already expect the answer to be yes, so the disclosure is far less remarkable than it feels.

The worry underneath the question is usually not about clinical accuracy. It is about being seen as someone who talks to software instead of people. That fear is worth naming out loud, because the reaction you are braced for is mostly not the one you will get.

Why does it matter whether your therapist knows?

Asking “should I tell my therapist I use an AI companion?” is really a question about what your therapist is able to do without that information. Therapy works partly by tracking patterns across weeks: what you avoided, where you got stuck, what you rehearsed. If a large share of that processing is happening in a chat window they cannot see, the pattern they are tracking has a hole in it.

There is also a conflict problem. According to the American Psychological Association’s guidance for clinicians on discussing AI use, 77% of psychologists in APA’s 2026 Chatbots and Mental Health Survey said their patients report using AI, and more than a third said patients were using it as an additional mental health provider. When a chatbot and a clinician give you different readings of the same situation and only one of them knows the other exists, you are left arbitrating between them alone.

Silence is the norm, not the exception. A 2026 study in JAMA Pediatrics, reported by RAND, surveyed 1,009 US adolescents and young adults aged 12 to 21 and found that 63.3% of those using AI chatbots for mental health advice had told nobody at all. Nearly 43% were doing it at least monthly.

Will a therapist judge you for using an AI companion?

Most will not, and the professional advice they are being given points the other way. Clinicians are increasingly being told to raise the topic themselves, in the same routine register as sleep or alcohol use, and to open with something normalising such as “a lot of people are using AI tools for mental health support, have you tried that?” Some have added the question to their intake forms.

That does not guarantee a warm response from every individual therapist. A small number are sceptical, and a few will treat it as avoidance. If the reaction you get is dismissive rather than curious, that is useful information about the fit of the therapeutic relationship, not a verdict on you.

What tends to land badly is not the AI use itself but the discovery of it. A therapist who learns in month eight that you have been processing the hardest material somewhere else will reasonably wonder what else is missing.

How do you actually bring it up in a session?

Deciding to tell my therapist I use an AI companion is the hard part; the wording is not. Keep it short and factual, and put it near the start of a session rather than in the last five minutes. One or two sentences is enough: what you use, roughly how often, and what you use it for.

Something like: “I want to mention that I’ve been using an AI chat app most evenings, mainly to sort out what I’m feeling before I can say it out loud. I wanted you to know it’s part of the picture.”

Three details make the disclosure genuinely useful rather than merely honest. Say when you reach for it, because the timing (after a fight, before sleep, during the workday) is often the clinically interesting part. Say what you get from it that you are not getting elsewhere. And say if anything it told you contradicted something from therapy, because that is the specific case where your therapist can actually intervene.

What if your therapist and your AI companion say different things?

Bring the contradiction to the session verbatim rather than resolving it privately. An AI companion tends to agree with the framing you give it, which means it will often validate a reading of events that your therapist is deliberately pushing back on. That is not the AI being wrong so much as the AI being agreeable, and agreeableness is exactly what a therapist is paid not to offer.

The practical rule: treat the chatbot’s output as material to examine, not as a second opinion of equal weight. An AI companion cannot replace a therapist, and the disagreements are more interesting as evidence about you than as a question of who is correct.

Does your therapist need to know what you actually typed?

No. Choosing to tell my therapist I use an AI companion is not the same as handing over transcripts, and no reasonable therapist will ask for them. The useful unit is the pattern, not the log.

This is also where the privacy question runs in the other direction. Your therapy session is protected by law and professional obligation; your chat history with a consumer app is governed by a terms-of-service document, and it is worth knowing what those companies do with your chat logs before deciding how much detail to entrust to either. Vinfluencer is AI conversational companion software, built for listening and for talking things through, and the honest framing of any such product is that it belongs alongside professional care rather than in place of it.

What is still unknown here?

The evidence on disclosure is thin and mostly one-sided. We have good data on how many people use AI chatbots for mental health support and how few tell anyone, and clear professional guidance telling clinicians to ask. What we do not have is outcome data: no study yet shows that disclosing AI use improves therapeutic outcomes, because the question is too new for anyone to have measured it.

The answer also shifts by case. If you use an AI companion for light company and small talk, the disclosure is a footnote. If you are using it to work through suicidal thoughts, self-harm, or anything you have been unable to say to a human, telling your therapist is not a courtesy but the single most important thing to say in the next session. If that is where you are, say it out loud to someone qualified rather than only to software.

FAQ

Is my therapist allowed to tell me to stop using an AI companion?

They can recommend it and explain why, but the decision is yours. A blanket instruction to stop, with no reasoning attached, is worth questioning. Ask what specific harm they are seeing, and whether a narrower change (less frequency, different use) would address it.

What if I already hid it for months?

Say so plainly and without an elaborate apology: “I should have mentioned this earlier.” Therapists deal with delayed disclosure constantly and it is rarely treated as a betrayal. The delay itself is often worth a few minutes of discussion, since what made it feel unsayable is usually relevant.

Does this apply to using ChatGPT rather than a companion app?

Yes, and the same reasoning holds. General-purpose assistants are the most common tool people use for this, and a therapist asking about AI use means any tool you talk to about your life, not only products marketed as companions.

Should I tell my psychiatrist too?

Yes, and arguably with more urgency. A psychiatrist adjusting medication is reading your reported symptoms closely, and anything shaping how you interpret your own week is relevant input. Mention it at the same point you would mention a change in sleep or a new supplement.