AI transparency
AI assists. A human stays responsible.
AI may assist with drafting and organization. A human coach reviews final plans and safety decisions.
Where AI may help
Organizing a redacted coaching brief, drafting a conservative plan after founder triage, and performing a separate structured safety review.
Where AI does not decide
Medical suitability, diagnosis, treatment, emergency guidance, final safety approval, or whether a plan is delivered.
Fail loud
If a configured model fails, the workflow stops. It does not silently switch to a different model or automatically deliver.