How our content is made
Last updated 15 August 2026
We use AI to build parts of this curriculum. We think you are entitled to know exactly where, exactly how, and exactly what a human being is responsible for. This page says so plainly, and we will keep it current as the method changes.
Why we use it at all
The reason is distribution, not cost-cutting. World-class trauma knowledge exists; it does not reach most of the surgeons who need it. AI is what makes one filmed curriculum available in many languages, keeps it current when guidelines change, and renders anatomy that no camera in a real operating theatre can show — the plane beneath the surface, the vessel at risk, the decision ahead.
What is filmed, and what is rendered
- Real operative footage is the foundation of the curriculum. Surgery is learned by watching surgery.
- Rendered sequences are used where a camera cannot show what a learner needs to see, or where a real recording could not be used without risk to a patient's privacy. Renders are always labelled as renders. We do not present a rendering as real surgery.
- Narration and expert commentary are recorded by real senior surgeons. Where a faculty member's voice is reproduced in another language, it is done under a licence they have signed, and it is disclosed.
What a human being is responsible for
Every module is authored from primary evidence and carries an evidence table mapping each clinical claim to its source. Every sequence, filmed or rendered, is reviewed and signed by a named faculty member before release. Nothing publishes without that signature.
The judgement is never artificial. A model does not decide what is clinically correct here; a named surgeon does, and their name is attached to it.
Patient privacy
Consent governs what enters our pipeline, not only what leaves it. Footage is only used as source material where there is a documented lawful basis for that use. Material without one is not used. Where we need clean source material, we film purpose-built simulation rather than reach for an archive we cannot account for.
No identifiable patient data is uploaded to any third-party AI service, under any circumstances. Our vendors operate under enterprise terms that prohibit training on our inputs.
TraumaOne is pre-launch. This page describes the method we are building to and hold ourselves to, not a catalogue of published modules. As the first modules are released we will update this page with the specifics of how each was produced.
Questions
If you are a clinician, an accreditor or an institution and you want more detail than this page gives, write to us at hello@traumaone.academy and we will answer directly.
← Back to TraumaOne