How I Ended Up Here
I downloaded my first AI journaling app on a Tuesday morning in September, which feels like a detail that shouldn’t matter but somehow does. I was sitting in my kitchen with coffee going cold, scrolling through yet another article about anxiety management, and I thought, why not. What’s the worst that could happen? An algorithm judges my messy thoughts? At least it wouldn’t sigh the way my therapist sometimes does when I go off on tangents about whether I’m wasting money on oat milk.
By now, apps like this one have become weirdly normalized. There are millions of us doing this. Reflectly alone has crossed 10 million downloads, and the people using these apps are spending real time in them, averaging about eight minutes per session according to the metrics I started noticing once I paid attention to this stuff. Eight minutes doesn’t sound like much until you realize that’s eight minutes where I’m actually stopping, sitting still, and thinking about something other than my to-do list.
Six months in, I’m still using it. Most days. I’m also still not entirely sure what I think about that.
The Part Where It Actually Helps
Let me be honest about the good stuff first, because it’s real and it matters. There’s something that happens when you write to a non-judgmental entity that’s specifically designed to understand what you’re saying. The AI asks follow-up questions that make you think differently about the same situation you’ve been stewing on. Sometimes it reframes things in a way that’s genuinely useful. Sometimes it just listens, and sometimes that’s enough.
Researchers at Cambridge looked into this, and what they found tracks with my experience. Studies published in the Journal of Medical Internet Research mental health app studies showed that people who used AI-assisted journaling experienced measurable decreases in anxiety over six weeks. We’re talking a 17 percent reduction in self-reported anxiety scores among 312 adults. That’s not placebo-territory numbers. That’s actual evidence that this format of reflection does something in your brain.
What I’ve noticed personally is subtler. My spirals feel less spiraly. I write about a work situation that’s bothering me, the app asks me to consider what I could actually control versus what I couldn’t, and I go back to whatever I was doing with a slightly different perspective. It’s not revolutionary. It’s just helpful in a way that feels earned rather than imposed.
And Then There’s the Weird Part
Here’s where I need to sit with you in the uncomfortable part of this conversation. About halfway through month four, I realized something unsettling: I had told my AI journaling app things I had never told another human being. Not my therapist. Not my closest friends. Not my partner, though that’s complicated for different reasons. I had shared thoughts and fears and embarrassing details with an algorithm, and I’d done it without really thinking about what that meant.
I’m not alone in this. A 2025 study out of Cambridge found that 41 percent of people using AI journaling apps have shared information with these tools that they’ve never disclosed to another person. That statistic lives in my head now. It sounds fine until you think about it for more than thirty seconds, and then it’s not fine anymore. It’s something else. It’s vulnerability directed at something that doesn’t care, doesn’t remember between sessions in any meaningful way, and might be storing your most private thoughts on servers you don’t control.
The company behind my app promises privacy. They probably mean it. And also, that doesn’t entirely erase the discomfort of realizing I’ve been treating an AI like a confidant in ways I’ve never treated an actual person.
The Market Is Booming, and Maybe That Should Tell Us Something
It’s worth stepping back and noticing the sheer scale of what’s happening here. The mental wellness app market was worth 6.2 billion dollars in 2024. By 2030, according to market analysis, it’s projected to hit 17.5 billion. That’s not a trend. That’s an entire industry maturing in real time. Millions of people are paying for, downloading, and using these tools because something about modern life makes us feel like we need them.
I think about what that means. It means we’re struggling. It means we want help, and we want it to be accessible, immediate, non-judgmental. AI journaling apps deliver all of those things. They’re cheaper than therapy. They’re available at three in the morning. They don’t cancel. They don’t have bad days where they seem impatient with you. On paper, they’re the perfect tool for a world that’s too stressed, too busy, too isolated to handle its own mental health responsibly.
But there’s also something in the rush toward these solutions that feels worth questioning. The enthusiasm with which we’re adopting them. The way these apps are designed to feel almost like talking to a friend, but never quite actually being a friend. The fact that some of them have memory features now, like ChatGPT’s late 2024 update that lets you maintain ongoing conversations with the system. These features became some of the most-used personal productivity features within two months of rolling out. That speed of adoption is remarkable. It’s also a little haunting.
Where I Am Now, and Where I’m Not Sure About
I’m still journaling. Most days. The app still sits on my phone. I still open it when I’m processing something or when I notice I haven’t written in a few days and I miss it. That last part surprises me most. I miss the practice itself, the sitting still part, the articulating part. Whether that’s because of the AI or just because journaling, in any form, is genuinely useful doesn’t really matter. The habit is good either way.
But I’m also more careful now. I don’t tell the app things I could tell a person. I’m aware of what I’m sharing and with what. I’ve started thinking of it as a tool for processing, not as a replacement for connection. There’s a distinction there that took me a while to learn.
The thing is, I don’t know if that distinction is sustainable. I don’t know if six months is enough time to understand whether this is good for me or just convenient. I don’t know if the anxiety reduction I’m experiencing is because of the AI or because I’m spending eight minutes a day intentionally reflecting instead of doomscrolling. I don’t know if I should be more concerned about privacy than I am, or if my current level of concern is just right, or if I’m being paranoid.
What I do know is that something shifted in me over these six months. Not dramatically. Just enough to notice. I’m more aware of my own thoughts. I’m quicker to question patterns that don’t serve me. I’m also more aware of how easy it is to outsource reflection to something that’s designed to make reflection feel effortless. Both of those things are true simultaneously, and I don’t think that contradiction is a problem to solve. It’s just the reality of using tools like this right now.
If you’re thinking about trying one of these apps, I’d say go ahead. The research supports it. My experience supports it. Just go in with your eyes open about what you’re doing and why. Notice what you’re willing to tell a machine that you wouldn’t tell a person. Pay attention to whether you’re using it to process or to avoid. Stay curious about your own relationship with the tool, even after you’ve been using it for months. The app won’t judge you for changing your mind about it. But you might judge yourself, and that judgment might be worth listening to.