FAQ

Questions, answered honestly.

Short answers, each grounded in something you can verify elsewhere on this site. If something is not yet available, we say so.

Is this a real, working product or a mockup?

It is a real, working system. You can log in to the live demo and use it yourself — it runs on synthetic data with zero protected health information. The home page shows the automated test gate; you can run that suite during a pilot.

What does it cost?

We are onboarding a small group of design partners, and pricing is set together with each partner. There is no public price sheet yet, and we do not publish numbers we cannot stand behind. A pilot has no long-term commitment.

Is it HIPAA-compliant? Can it handle real PHI today?

No — and we are careful here. Today everything runs on synthetic, zero-PHI data, enforced by tests. There is no such thing as "HIPAA-certified software." Handling real data is a future, gated step that requires a signed Business Associate Agreement and a governed compliance boundary. We never market that as available today.

What is live right now versus on the roadmap?

Every capability on the site is tagged live, coming, or gated. The Live vs. roadmap page is the single map of exactly what is available today, what is coming, and what is gated behind a compliance step.

Who is it for?

Provider facilities, MSOs, SCN and CBO entities, consultants, EMRs, and billing companies in behavioral health — with a first focus on NY Medicaid and the SCN program.

How do you keep our data safe when we integrate?

External connectors ship disabled behind environment switches ("real-seam OFF"), the defaults are synthetic, and the PHI boundary is built to fail closed. The developer & integration docs walk through exactly what crosses the wire and map each security claim to something you can inspect.

Does the assistant (Sophie) use an external AI model?

Not today. Sophie is a deterministic, permission-aware local preview that answers only from synthetic, on-screen data and names its source — no large language model and no external AI calls. A real-LLM upgrade is a gated future step.

How do I try it?

Open the live demo and sign in with a synthetic role account, or request a design-partner pilot to walk one workflow end-to-end with our help.