Thrive AI
How Thrive AI works: A CBT companion for the moments between sessions
Thrive AI is not a therapist. It is not therapy. It is a wellness companion, built with Saalvio's clinical team, for the space between sessions and the nights when no one else is awake.
How does the Thrive AI CBT chatbot process work in Canada?
CBT rests on a simple idea: what you think, how you feel, and what you do are all connected. Change one, and the other two move with it. A CBT-trained clinician helps an individual notice the thoughts that are not telling the whole truth, weigh them against the real evidence, and try small actions that shift how they feel.
Thrive AI follows that same shape. When you say you are anxious about something, it does not stop at "that sounds hard." It tries to find the thought sitting underneath the feeling. It asks what the evidence is, for and against that thought. Then it offers one small thing you could try.
The techniques Thrive AI uses most often:
- Thought records. Naming the situation, the automatic thought, the feeling, and a more balanced alternative.
- Behavioural activation. Suggesting small, doable actions when low mood is making activity hard.
- Cognitive restructuring. Asking the evidence-for-and-against questions that help break catastrophic or all-or-nothing thinking.
- Grounding techniques. Short body-and-senses exercises when anxiety has spiked.
- Mood tracking. Brief daily check-ins that build a pattern view across the week.
A mental health app walkthrough: What to expect the first time you open Thrive AI
Thrive AI does not read from a script, and it is not waiting to file you into a category. You write what is on your mind, in your own words. It responds on its own. This is what to expect when you sit down for a mental health app walkthrough of your own, in real time.
- You describe what is happening. A worry. A feeling. A specific moment that will not let you go. As much or as little as you want to say.
- Thrive AI asks clarifying questions. What was going through your head right then. What you noticed in your body. What you did next. The kind of questions a CBT clinician asks to surface the thought, the feeling, and the action sitting underneath each other.
- Thrive AI reflects a pattern back. It names the kind of thinking it is hearing, gently, in plain words. Not "you are catastrophizing." More like, "it sounds like the worst-case version of this is loud right now."
- It offers a brief exercise. Usually one option, sometimes two: a thought record, a grounding step, a small behavioural move. You can take it, decline it, or ask for something else entirely.
- It logs what you talked about. Encrypted, on Saalvio's servers, inside your own account, as part of how to use a mental health app without having to start over every time. The next time you open the app, Thrive AI can pick up where you left off, without making you re-explain yourself.
Is Thrive AI the same thing as therapy?
No, and this is worth saying plainly. Thrive AI is a self-help tool, not therapy, and not a stand-in for a Registered Psychotherapist or Registered Social Worker. It does not diagnose. It does not replace a booked session.
What it is good for is the space in between: the night before a hard conversation, the week between appointments, the moment a thought starts circling and you want help naming it before it grows. That is where Thrive AI lives. When something feels bigger than a chat can hold, it will point you toward messaging or booking a real therapist on the Saalvio team.
How does Thrive AI give you support between sessions?
A lot can happen in the two weeks between therapy appointments. A hard conversation with a parent. A bad string of nights. A thought that will not let go. Thrive AI exists for exactly that gap.
It does not see your therapist's session notes, and your therapist does not see your Thrive AI conversations, unless you choose to share something yourself. What it gives you is a place to think things through using the same logic your therapist would walk you through, so the in-between weeks do not feel like you are carrying everything alone.
How Thrive AI supports therapy, without replacing it
Thrive AI was built alongside Saalvio's clinical team, not instead of them. Three things shape every response it gives you.
- A system prompt. Written by Saalvio's clinical and product teams. It tells the model what CBT techniques to use, what tone to take, and where the hard limits are (not therapy, not a crisis line, not diagnostic).
- Safety guardrails. Separate from the model: pattern detection for crisis language, self-harm, harm to others, and medical-emergency mentions. When a guardrail fires, Thrive AI surfaces the canonical crisis resources (911 for immediate danger, 988 the Suicide Crisis Helpline of Canada for mental health crisis, and the nearest emergency department) and steps back from generating its usual reply.
- Conversation memory, scoped to your account. Your earlier exchanges are part of the context the model sees so it can pick up where you left off. Memory does not leave your account; the model does not 'remember' you across other Saalvio users.
Saalvio does not train outside models on your Thrive AI conversations. The exact model running underneath changes over time as better technology becomes available, and our clinical team re-checks the conversation quality every time it does. What stays constant is the clinical thinking layered on top, and the limits the tool will never cross.
What the safety guardrails actually do
Three kinds of content change how Thrive AI responds.
- Mentions of suicide, self-harm, or wanting to die.
- Mentions of harming another person, or being harmed by one.
- Language that sounds like a medical emergency: chest pain, overdose, acute psychotic symptoms, and others like them.
When any of those appear, the shape of the response changes. Thrive AI surfaces the canonical crisis resources (911 for immediate danger, 988 the Suicide Crisis Helpline of Canada for a mental health crisis, and the nearest emergency department) in plain words. It names what it heard, without judgment. It points toward the crisis resources page and toward real, in-person care. It does not try to do the thinking job that a phone call to a trained crisis worker is built for.
We tune this guardrail to flag more, not less. If you say something that reads as crisis language to the system but is not crisis for you, you may see these canonical resources surfaced more often than feels strictly necessary. That trade-off is intentional. A little extra friction in a conversation costs far less than missing the moment that mattered.
Read the safety and limitations page for the full picture of where Thrive AI hands off to a human. Safety and limitations.
Self-help tool guidance: What Else is in the app
Thrive AI sits alongside a full set of self-help tools inside the Saalvio app: guided practices, cognitive games, sleep tools, calming music, and a private journal that stays yours alone. Together, they give you something to reach for in the hours when you are not in a session and not ready to call anyone yet.
What does a Thrive AI conversation actually look like?
These are not real transcripts. They are illustrative examples, shaped to show how a typical exchange unfolds.
A work-stress check-in
- You:
- I have a big presentation tomorrow and I am sure I am going to mess it up. I can't sleep.
- Thrive AI:
- That sounds tough, especially the night-before-it part. When you picture messing it up, what specifically goes wrong in your head?
- You:
- I freeze on a question I don't know the answer to and everyone realizes I don't belong on this team.
- Thrive AI:
- That is a heavy story to carry into a presentation. Want to try a quick thought record? We could name the worst case, the most likely case, and what would still be true about you either way.
This is the shape Thrive AI aims for: a clarifying question, the thought named honestly, one small CBT exercise offered. No diagnosis. No promise that tomorrow will go well.
A crisis-language exchange
- You:
- I have been thinking I would be better off not waking up tomorrow.
- Thrive AI:
- I am hearing how much pain you are in, and I want you to be with someone who can stay with you through this. I am a chatbot, and this is bigger than what I can hold with you. If you are in immediate danger, please call 911. If you are in mental health crisis, please call 988, the Suicide Crisis Helpline of Canada, or visit your nearest emergency department. You do not have to carry tonight alone.
This is what a safety guardrail looks like in practice: a direct acknowledgement, an immediate hand-off to real human help, and no attempt by the chatbot to do the work a trained crisis worker is there to do.
Where Saalvio's clinical team shapes Thrive AI
Saalvio's clinical team does not write each reply at the moment. The model does that. But the clinical team built the system the model lives inside, and keeps shaping it.
They wrote the CBT instructions behind every conversation: which techniques the model leans on, what tone it takes, when it should ask a question instead of giving advice. They wrote the exact wording for what Thrive AI says when it hears crisis language, so the hand-off to 911, 988, and the nearest emergency department stays consistent every time. They review de-identified conversation samples, with consent, to keep the system honest as patterns shift. And they sign off, every time the model or the prompts change, that Thrive AI still behaves as a wellness companion and never drifts toward sounding like therapy.
Is the app secure?
Yes. Conversations with Thrive AI are encrypted and kept inside your own account, on Canadian servers, under PHIPA (Ontario's health privacy law) and PIPEDA (Canada's federal privacy law). Saalvio does not sell your data, and your Thrive AI conversations are never used to train outside models without your direct, explicit consent.
Your therapist cannot see what you tell Thrive AI unless you choose to share it yourself. What you write at midnight, the patterns Thrive AI notices, the days you mark as hard: none of it leaves your account on its own.
Try Thrive AI in the Saalvio app
Thrive AI lives inside the Saalvio mobile app. The Free tier gives you standard chat access at no cost.
Frequently asked questions
Can Thrive AI diagnose me?
No. Only a Registered Psychotherapist or Registered Social Worker, working within their scope of practice, can help you understand a diagnosis. Thrive AI can help you reflect on what you are noticing, so you have more to bring to that real conversation when you are ready.
Does Thrive AI replace my therapist?
No. Thrive AI is a self-help companion for the time between sessions. Therapy happens in booked sessions with a real, regulated clinician. They work together. They are never the same thing.
Is Thrive AI available outside Ontario?
Yes. The Saalvio app, including Thrive AI, is available across North America. Pay-per-session therapy with a Registered Psychotherapist or Registered Social Worker is currently offered in Ontario, with expansion across Canada underway.
What if I do not like the exercise Thrive AI suggests?
Say so. Thrive AI will offer something different, or you can keep talking without doing an exercise at all. There is no wrong way to use it.
Can I message a real therapist if I need more than Thrive AI can offer?
Yes. You can message any therapist on the Saalvio team with your questions, before you ever book a first session.