Can an AI companion help with depression?

Partly, and only around the edges. Structured therapy chatbots show small measured drops in depressive symptoms. The claim that an AI companion helps with depression is on much shakier ground. Companion apps were never built as treatment, and the longest-running study of them found mixed results, including more distressed language over time. As an easy place to put words to a bad day, a companion can help. As treatment? No.

What matters here is that two different products keep getting lumped together. A therapy chatbot walks you through cognitive behavioural exercises from a script. An AI companion is open-ended conversation software with a persona. Almost all the encouraging research is about the first one.

What does the research actually show about chatbots and depression?

Research on chatbots and depression is further along than most people think, and more modest than most headlines make it sound. A 2025 meta-analysis of 38 randomised trials covering 7,401 participants found chatbots produced a statistically significant drop in depressive symptoms, with a standardised effect size of g = 0.31. That’s a small-to-moderate effect. It was bigger in clinical and subclinical samples than in people who weren’t depressed to start with.

A separate systematic review of generative AI mental health chatbots in the Journal of Medical Internet Research found a similar effect size of 0.30 across 6,314 participants.

Two caveats come with those numbers. Most trials lasted weeks, not months. And most tested purpose-built therapy tools designed with clinical input, not consumer companion apps.

Is a companion app the same thing as a therapy chatbot?

A companion app isn’t the same thing as a therapy chatbot, and assuming the evidence carries over is the single most common mistake people make on this topic. Therapy chatbots like Woebot or Wysa were designed around specific protocols, tested in trials, and built to steer the conversation toward an exercise. Companion software is built to keep talking.

That changes what the software does when you’re low. A therapy tool is designed to break a thought loop. A companion is designed to stay in the conversation with you. Those are different goals, and sometimes opposite ones.

So can an AI companion help with depression? Honestly, almost nobody has tested that directly. The evidence people usually quote belongs to the product category next door.

Why might an AI companion make depression worse for some people?

An AI companion can make depression worse in three documented ways, and none of them need the software to break. The most careful evidence comes from a 2025 quasi-experimental study of nearly 2,000 users that compared their public Reddit posts for a year before and a year after they started using a companion app. Emotional expression and focus on other people went up. So did language about loneliness and suicidal ideation.

Risk one is displacement. Support that feels like enough can quietly replace the harder step of calling a friend or booking an appointment, and depression already hands you plenty of reasons to skip both.

Risk two is agreement. Companion models tend to agree with you too readily. That’s exactly backwards for depression, where what helps is often a gentle challenge to a distorted belief, not validation.

Risk three is availability. Something that’s always awake at 3am takes away the natural friction that would otherwise end a rumination spiral.

When is talking to an AI companion a reasonable thing to do?

Talking to an AI companion is reasonable when you know what you’re using it for. Naming a feeling at all is a documented first step, and for lots of people a blank chat window is easier to face than someone’s expression. Low-stakes practice at saying what’s wrong is a real benefit. It just isn’t a clinical one.

Good uses are narrow: rehearsing what you want to tell your doctor, having company on an evening that would otherwise be silent, or spotting a pattern across weeks of your own words. Vinfluencer.ai is AI conversational companion software, and the same honest framing applies to it: a place to think out loud, not a course of treatment.

Bad uses are the reverse: using the chat so you don’t have to tell a human you’re struggling, or asking it to judge how bad things are.

What should someone with depression do alongside or instead?

Someone with depression should get treatment that’s actually been tested on depression, with conversational software as an extra at most. Depression is treatable. The treatments with decades of evidence are psychotherapy, medication where it’s indicated, and the boring basics of sleep, movement and daylight. A chatbot’s g = 0.31 is well below what those achieve.

Professional bodies including the American Psychological Association have consistently said AI chatbots should be treated as experimental or supportive tools, not substitutes for care. A companion can’t replace a therapist. It can’t adjust a medication. It can’t notice you’ve stopped leaving the house.

If your mood has been low for more than two weeks, or you’re having thoughts of harming yourself, please talk to a doctor or a crisis line, not an app. That’s not a legal disclaimer. It’s the part of the answer that matters most.

What is still unknown about AI companions and depression?

A lot is still unknown about AI companions and depression, and anyone saying otherwise is getting ahead of the evidence. No randomised controlled trial has tested a consumer companion app as a depression intervention. The best long-term evidence is observational and based on public Reddit language, not clinical assessment. So it can’t tell us whether companion use made anyone’s depression worse, or whether people already sliding downhill went looking for the software.

Which way causation runs is the open question across this whole literature. Depression predicts companion use. Companion use predicts distressed language. Current study designs can’t pull those apart.

Nobody has mapped dose either. We don’t know whether twenty minutes a day and four hours a day are different in kind or just in degree, and the answer probably varies by person in ways research can’t sort out yet.

Frequently asked questions

Are AI companions approved for treating depression?

No AI companion app is approved by the FDA or any equivalent regulator to treat depression. A handful of purpose-built digital therapeutics have regulatory clearance. Consumer companion apps are general-purpose software sold as wellness or entertainment products, which puts them completely outside medical device oversight.

Is an AI companion better than nothing when no therapist is available?

Maybe, for mild low mood, and with waiting lists that’s a real situation. The evidence behind that hope comes from structured therapy chatbots, not companion apps, though. And “better than nothing” stops being true the moment the app becomes the reason you never get on the waiting list.

Can an AI companion tell if I am getting worse?

Not reliably. Companion software has no clinical assessment built in, can’t see your sleep, appetite or behaviour, and can’t contact anyone for you. Most major apps now send explicit crisis language to helpline information. That’s a safety feature, not a way of detecting worsening depression.

Does using an AI companion for depression mean something is wrong with me?

No. Turning to a private, available, judgement-free place to put hard thoughts is a normal response to feeling bad, and surveys keep finding emotional support among the top reasons people use these apps. Wanting more than software is normal too. Act on it.