ElderWatch AIBack to home

AI Monitoring Consent

ElderWatch AI — AI Monitoring Consent

Version 1.0.0 — Effective Date: [INSERT DATE]

What This Consent Covers

This document explains, in plain language, how ElderWatch AI's artificial intelligence system processes behavioral data, what it can and cannot do, and what you're agreeing to when you check this box.

How the AI Works

  1. Learning phase (first 14 days): The AI observes patterns — when the Elder User typically wakes up, how active they are, when they use their phone, when they charge it, how often they make calls, and which places they regularly visit. No alerts are generated based on anomalies during this phase; the system is establishing what "normal" looks like.

  2. Ongoing monitoring: After the learning phase, the AI compares daily activity against the established baseline and calculates an anomaly score (0–100) and risk level (Green/Yellow/Orange/Red).

  3. Alerting: When the risk level rises above Green, designated Family Caregivers are notified. Sustained Orange/Red levels trigger the escalation sequence (push notification → repeated push → SMS → phone call → emergency contacts).

What the AI Is — and Is Not

The AI is:

  • A statistical pattern-recognition system
  • Designed to flag deviations, not diagnose conditions
  • Continuously refining its baseline as more data is collected

The AI is NOT:

  • A medical diagnostic tool
  • Capable of detecting every health emergency (e.g., a silent heart attack with no behavioral change beforehand may not trigger an alert)
  • A replacement for regular check-ins, medical care, or emergency services
  • Free from false positives (unusual but harmless days, like a holiday or hospital visit, may trigger alerts) or false negatives (some real emergencies may not produce a detectable behavioral signature)

Data Used by the AI

  • Device unlock timestamps
  • Movement/step data from device sensors
  • Charging event timestamps
  • Call frequency (not content)
  • Location clusters (not continuous tracking)

Human Oversight

No ElderWatch employee manually reviews raw behavioral data as part of normal operations. Aggregated, de-identified data may be reviewed by our engineering team solely to improve model accuracy. You may request a data processing audit by contacting privacy@elderwatch.ai.

Your Right to Withdraw

You (or the Elder User, if monitoring was set up on their behalf) may withdraw AI monitoring consent at any time:

  • Family Caregiver: Settings → Privacy → Disable Behavior Learning
  • Elder User: Ask your Family Caregiver to disable monitoring, or contact support directly

Withdrawing consent stops new data collection but does not retroactively delete already-processed baseline data unless you separately request data deletion under our Privacy Policy.

Acknowledgment

By accepting this consent, you confirm that:

☐ You understand the AI provides probabilistic alerts, not certainties or diagnoses ☐ You understand false positives and false negatives are possible ☐ You will not rely on ElderWatch AI as a sole means of ensuring the Elder User's safety ☐ You have explained this system to the Elder User in terms they understand, or hold valid legal authority to consent on their behalf

This consent works alongside the Terms & Conditions, Privacy Policy, and Data Processing Agreement. In case of conflict, the most protective term for the Elder User's privacy shall prevail, subject to final legal review.