What if your
second opinion
was always available?
OsteoAI is a clinical reasoning assistant built for osteopaths. Paste a presentation — age, complaint, findings, history — pick a mode, and reason through it with a colleague who speaks your framework. Five modes: Clinical Reasoning, Differential Dx, Treatment Plan, Case Simulation, General.
Every case builds a living knowledge graph.
As you reason through a case, every concept the model leans on becomes a node. Every clinical connection becomes an edge. The graph builds in real time and persists across the session — so you can see what was derived from what, drag nodes to explore relationships, and watch new connections emerge as the conversation deepens.
The graph doesn't hide what it doesn't know. Uncertainty renders differently from confidence. The structure itself is the reasoning made visible.
Available in Clinical Reasoning and Treatment Plan modes. Open it before the first message to watch it build from the start.
Driver
The upstream cause. Dense, filled node with a full dendrite bloom. Removing it resolves the dysfunction. Treating anything else gives temporary relief only.
Amplifier
Sustains or worsens the pattern without starting it. Amber. Comorbidities, psychosocial load, sleep disorder — anything that raises the clinical volume.
Compensation
Adapting to the driver. Outlined node, fewer dendrites. Treating it gives temporary relief only — it returns until the driver is addressed.
Pathway
Teal with a dashed outer ring. The nerve, fascial chain, or reflex arc through which the driver acts. Not a treatment target — name the source instead.
Uncertain
Dashed outline. The graph admits when it isn't sure. A differential that needs more information before it can be classified. Not false confidence.
Built to know its own limits.
The hard part of a clinical assistant isn't producing an answer. It's knowing when the answer shouldn't be trusted — and saying so before you act on it.
Red flags come first
Cardiac, systemic and neurological screening runs on every case and re-runs on every follow-up. A flag raised in turn one doesn't quietly disappear by turn four — and a referral is never suggested without naming who, and how urgently.
It tells you when it's guessing
When a presentation falls outside what the knowledge base covers well, the response says so up front — before the clinical content, not buried after it. Thin evidence produces hedged reasoning and named open questions, not confident-sounding filler.
Written by a practising osteopath
The knowledge base grows from real clinical encounters — the findings that actually differentiate one presentation from another, and the mistakes worth not repeating. Every gap a case exposes becomes the next thing written.
What this is — and isn't.
A reasoning peer, on call.
- A second opinion at 11pm, before the next morning's appointment.
- A study partner for students running through case sims and differentials.
- Built for osteopaths. Every mode reflects how osteopathic practitioners actually reason through a case.
- An honest assistant. Renders uncertainty as uncertainty — and says outright when a presentation sits outside what it covers well.
- Safety-screened by default. Red flags, referral urgency and destination are checked on every case, not offered as an afterthought.
Not a diagnosis. Not a regulator.
- Not a medical device. Not CE-marked, not FDA-cleared. Educational support only.
- Not for patients. The voice is peer-to-peer; outputs aren't safe for direct patient consumption.
- Not a replacement. For your training, your judgment, or hands-on assessment.
- Not a black box. If we can't show the chain, we don't show the conclusion.
- Not complete. Coverage is deeper in some regions than others — and the assistant will tell you which case you're in.
Pricing, shortly.
Rates are being finalised ahead of general release. Beta members will be told before anything is charged — and whatever you sign up at is what you keep.
The whole assistant.
All five modes. Knowledge graph. Full monthly query allowance. Beta rate locked permanently once announced.
- Free to use during beta. 20 queries per month, no card required.
- All five modes — Clinical Reasoning, Differential Dx, Treatment Plan, Case Simulation, General.
- Knowledge graph — builds visually with every case.
- Beta rate locked — whatever you sign up at is what you pay, permanently.
- A reduced student rate will be available with a verified institutional email.
- No surprise billing. Nothing is charged without notice and your explicit sign-up.
For the whole team.
Shared query pool across the practice. All five modes for every practitioner on the account.
Built for education.
Osteopathy schools, university departments, larger clinic groups. Custom limits, analytics, priority support.
Asked often.
If yours isn't here, write to us. We answer ourselves.
What model is behind it?
A large language model with domain-specific reasoning architecture and a curated osteopathic knowledge base. The model is guided by clinical reasoning rules built with practising osteopaths — not general medical prompting.
Is patient data safe?
Don't paste it. The product is built for de-identified presentations. We don't ingest, store, or train on patient records. If you need a PHI-compliant deployment, talk to us about a private instance.
What does it cost?
Nothing during beta — 20 queries a month, no card. Paid rates are being finalised and will be announced before anything is charged. A reduced student rate with a verified institutional email is part of the plan.
Can students use it?
Yes — and we built Case Simulation mode largely for you. A reduced student rate will be available with a verified institutional email, and if your school takes an institutional licence, access may be included.
What happens when beta ends?
Beta members get notice before anything changes, and the rate you sign up at holds for as long as your subscription is continuous. We expect to add tiers — early clinicians stay on the price they started at.
What happens if it doesn't know?
It says so. Coverage is deeper in some regions than others, and when a presentation falls outside what the knowledge base handles well, the response opens by telling you that — then reasons from general principles and names the specific open question, rather than producing confident-sounding filler.
How is this different from a general AI chatbot?
Domain. Voice. Mode-specific prompting. A knowledge graph that builds with every case. Red flag screening that persists across a conversation rather than firing once. And a deliberate refusal to do things outside scope — it won't diagnose, won't recommend specific medication, won't write patient-facing copy.
Practised by
practitioners.
OsteoAI is built in active dialogue with clinicians. The product gets sharper every time a case reveals a gap in the reasoning.
Founding practitioners
Beta members shape what gets built next. If your caseload reveals a reasoning gap, that gap becomes a fix — usually within days.
Locked-in pricing
Whatever rate you join at holds for the life of your subscription. No re-pricing when we leave beta. You backed it early — that counts.
Word of mouth
OsteoAI grows the way good clinical tools always have — one colleague recommending it to another.
Try a case you
already know.
20 free queries, no card required. All five modes from the first session. If the reasoning doesn't hold up against your clinical judgment — you lose nothing.
What case are we
looking at
today?
20 free queries. No card. All five modes from the first session. Open the assistant and try a case you already know the shape of.