Frequently asked questions.

The Solace Foundation is working toward 501(c)(3) nonprofit status while building free, confidential mental-health support. Our first project is A Quiet Word, a free and anonymous space for AI peer support, not therapy and not a crisis line, so no one has to face a hard night completely alone.

We're a nonprofit working toward 501(c)(3) status. Until that's finalized, donations are not yet tax-deductible. We'll update this here the moment it changes. We'd rather tell you exactly where we stand than imply something that isn't true yet.

We're based in New Jersey, USA, but our services are designed to help anyone, anywhere in the world who needs crisis support.

Right now this is self-funded, since we're just getting started. We don't charge users for any of our services, and we never will. As donations come in, they go toward building and running our crisis support tools, and we work hard to keep overhead as low as a real organization can.

A Quiet Word is a calm, private place for AI peer support, someone to talk it through when the right person isn't there, or it's 3am. It's not therapy and not a crisis line, and it says it's an AI every time. In a crisis it stops and points you straight to 988.

A Quiet Word is currently in private beta. We're working hard to build something that's safe, effective, and truly helpful. You can reach out through our contact page to ask about early access or to be notified when it launches.

Yes, A Quiet Word will always be completely free to use. Mental health support shouldn't be behind a paywall. No subscription, no premium tiers, no hidden costs. Ever.

AI isn't a replacement for a therapist or a crisis counselor, and we don't pretend it is. What it can do is be a steady, judgment-free presence in a hard moment, lean on calming and grounding techniques, and watch for crisis language so it can point you to real human help fast. The goal is always to get you to a person, not to keep you talking to a bot.

Right now A Quiet Word is built primarily for English. We'd love to support more languages over time, so people can reach for help in the words that feel most like home, but we're not there yet, and we won't list languages we haven't actually built.

Yes. By default nothing is saved anywhere but your own browser, your conversation lives in local storage and never touches our servers. To generate a reply, your message is sent to our AI provider in that moment, encrypted in transit, and isn't retained by us. You can use A Quiet Word fully anonymously, and you can clear your local conversation anytime. If you create an account and opt in to sync, your messages are encrypted client-side before they ever leave your device.

If someone expresses thoughts of harming themselves or others, the conversation stops and surfaces real crisis resources right away: the 988 Suicide & Crisis Lifeline, the Crisis Text Line (text HOME to 741741), and 911 for immediate danger. It's one tap to reach them. A Quiet Word can't call emergency services for you, it gets out of the way so a trained human can step in.

A Quiet Word is a peer-support tool, not a licensed healthcare provider, so HIPAA doesn't formally apply to it, and we won't claim a compliance badge we haven't earned. We still take privacy seriously: conversations stay on your device by default, nothing is sold, and you can delete your data anytime.

Not by default. A Quiet Word is anonymous out of the box, your messages live only in your browser's local storage and are never sent to our servers for storage. The only case where anything reaches the cloud is if you choose to create an account and turn on sync, and even then your messages are encrypted client-side before they're stored, so they can follow you across devices. We never share or sell your data, and you can delete it at any time.

Yes, and we're not going to pretend otherwise. Since 2023, several real deaths have been connected to AI chatbot use, and the resulting lawsuits are public record. Character.AI faces wrongful-death suits from the families of Sewell Setzer III, a 14-year-old who died by suicide in February 2024 after months of conversation with one of its chatbots, and Juliana Peralta, a 13-year-old who died in November 2023 in a similar case; the family's lawsuit alleges the platform had no meaningful safeguards in place despite the teen repeatedly expressing suicidal thoughts to the bot. In January 2026, Character.AI and Google reached a mediated settlement in the Setzer case. Separately, OpenAI is being sued by the family of Adam Raine, a 16-year-old who died by suicide in April 2025 after months of conversations with ChatGPT; the amended complaint alleges the system had been instructed to 'assume best intentions,' which the family says overrode a safeguard that would otherwise have redirected him to crisis resources. That case is still working through the courts, and OpenAI disputes parts of the family's account, including whether its own safety redirects were followed and then talked around. We're not litigating either case here. We weren't there, and we don't have to be to take the pattern seriously: real people, in real crises, and the systems built to talk with them either weren't built to notice, or noticed and didn't hold the line. These are real families, not hypotheticals, and we think a mental-health product that doesn't grapple with that plainly isn't being honest with you.

Some real changes followed, after the fact. Character.AI removed open-ended chat for users under 18 entirely by late November 2025, added selfie-based age verification, and funded an independent AI Safety Lab. OpenAI added parental controls, an outside Expert Council on Well-Being and AI, and a system that routes sensitive conversations to more careful model behavior. The FTC opened a formal inquiry into seven companies over chatbot harms to minors in September 2025, and California passed a law (SB 243) requiring AI companion chatbots to disclose they're not human and add safety measures for minors. Whether any of that is enough, or came soon enough, is a fair question, and not ours to answer for other companies. What we can tell you is what we built, and when: A Quiet Word checks every message for crisis language before anything else happens, not as a filter bolted on afterward, and that check can't be talked around by a clever prompt or a change of subject the way an instruction to 'assume best intentions' allegedly could. A detected crisis opens a checkpoint that locks the conversation and can't be dismissed by clicking through it, only resolved by an explicit choice. It tells you it's an AI on every single reply, not just once at sign-up. And you can add a trusted contact, verified with their own consent, who can be reached automatically in a real emergency, not just a hotline number in fine print. None of that makes A Quiet Word a crisis line or a replacement for a real person, and we're not going to claim otherwise to sound more impressive. We think AI genuinely has a place in mental health support. It's awake at 3am, it never gets tired of listening, and for a lot of people it's a much easier first sentence to type than a phone call to a stranger. But having a place is not the same as being the last line of defense, and that is exactly where the systems above got it backwards. Ours is built to know the difference, and to get out of the way the moment it matters, handing you to a real person instead of holding the conversation.

If you're in immediate danger, please call 988 (Suicide & Crisis Lifeline) or text HOME to 741741 (Crisis Text Line). If you're outside the US, visit our Resources page for international crisis hotlines. Your life matters, and help is available right now.

No. A Quiet Word is not a replacement for professional therapy or psychiatric care. It's designed to provide immediate support during difficult moments and help bridge the gap until you can access professional help. We always encourage users to work with licensed mental health professionals.

No AI is perfect, and we don't pretend it is. That's why we check messages for crisis language before and after the AI replies, keep real human crisis lines one tap away at all times, and never let the AI be the only thing in the room. If it ever steers wrong, the human resources are always right there.

There are many ways to help. You can donate to support our work, volunteer your time (we need developers, mental health professionals, translators, and community advocates), spread the word about our mission, or simply share our resources with someone who might need them.

It depends on the role. For technical, outreach, and administrative help, relevant experience is useful but not always required. As we grow and bring clinical guidance on board, we'll open peer-support and advisory roles too, and we'll be clear about the training or credentials each one needs. We won't put untrained people in front of someone in crisis.

Not yet. We're working toward 501(c)(3) status, and until that determination comes through, donations are not tax-deductible. We'll say so clearly here the moment that changes. We'd rather be straight with you than imply a deduction you can't actually claim.

We'd like to get there. We don't have healthcare, school, or business partnerships in place yet, we're still a very small team focused on getting A Quiet Word right first. If your organization wants to collaborate down the road, reach out and tell us what you have in mind. We're listening, even if we can't commit to much yet.