Responsible AI

ScribeMDPro is a documentation assistant, not a decision-maker. Clinicians remain the medical authority for every note.

Last reviewed: July 2026

Our operating principles

AI limitations

Speech recognition and language models can mishear, hallucinate, or omit detail. Every SOAP note is a draft that requires clinician review before it enters the medical record.

Human oversight

No note is auto-signed. The clinician edits, verifies, and takes clinical responsibility for the finalised document.

Bias mitigation

We test transcription and summarisation across accents, languages (including Yoruba, Hausa, and Swahili), and specialties, and iterate on prompts and provider routing to reduce systematic errors.

Transparency

The app surfaces per-segment confidence scores and timestamps so clinicians can inspect and correct low-confidence sections quickly.

Clinical responsibility

AI outputs are not a diagnosis, prescription, or treatment plan. Clinical judgment always overrides model output.

Appropriate use

ScribeMDPro is intended to support licensed healthcare professionals. It is not designed for direct-to-patient use, emergency triage, or autonomous clinical decision-making.

Not a replacement for clinical judgment

ScribeMDPro assists clinicians with documentation. It does not diagnose disease, prescribe medication, or replace the judgment of a licensed healthcare professional. Always confirm the final note reflects the encounter accurately before signing.