FDA · Digital Health · AI
Saami Salami Asl · Elisence

How Elisence thinks about FDA digital health principles and AI safety

This page provides a calm, high-level view of how Saami Salami Asl and the Elisence platform think about FDA-style principles for software safety, AI transparency, and responsible digital health design. Nothing here is medical advice or a regulatory claim; it simply explains how the platform is being shaped.

Good Machine Learning Practice Transparency & safety thinking Non-diagnostic posture

Purpose of this page: general transparency only. Not medical advice or regulatory positioning.

What the FDA focuses on

Overview

The US Food & Drug Administration (FDA) is concerned with safety, effectiveness, transparency and predictable behaviour of digital health tools. While Elisence is not a medical device, we believe in following the spirit of FDA-style thinking:

Clarity: Users understand what the system does and does not do.
Transparency: Decisions remain explainable.
Control: Features can be turned on/off through governance.

GMLP – Good Machine Learning Practice

AI

The FDA and other global regulators highlight “Good Machine Learning Practice” (GMLP). These ideas influence how Elisence is designed:

Data clarity: Inputs must be understood, not mysterious.
Explainability: Outputs must make sense, not behave randomly.
Human-in-the-loop: Humans should always remain in control.
Monitoring: Systems should be trackable and auditable.

Where Elisence sits

Positioning

Elisence is built for support, insight and structure, not medical diagnosis. Its role is to help families, clinicians and organisations talk about lifestyle and signals more clearly.

Non-diagnostic: No medical decisions are made by the platform.
Reviewable: Logic can be inspected by experts when needed.
Bounded: Behaviours follow strict guardrails.

Next steps

Info

As Elisence grows, elements of FDA-style governance—documentation, traceability, human oversight—will remain central. Any future regulatory pathways would follow the correct formal processes.