AI Transparency

How AI Is Used on NurtureArc

NurtureArc uses an AI language model to generate personalised wellness plans based on your assessment responses. This page explains how that works, what it can and cannot do, and what safeguards are in place.

The AI System

Your wellness report is generated by Anthropic's Claude model, accessed via a secure proxy. Claude is a large language model (LLM) trained on a broad corpus of text. It is not a medical system and has no clinical training.

What the AI Does

• Reads your pillar scores and question responses
• Identifies your areas of strength and areas for development
• Generates personalised action steps for each wellness pillar
• Produces a 90-day focus statement and closing message

What the AI Cannot Do

• Access your medical history, medication list, or risk factors
• Perform clinical assessment or diagnosis
• Replace the judgement of a healthcare professional
• Guarantee that its recommendations are appropriate for your circumstances

Limitations

AI language models can produce outputs that are plausible but incorrect, inconsistent, or inappropriate for specific users. Recommendations are generated fresh for each assessment and are not reviewed by a clinician before display.

How We Mitigate Risk

• The system prompt instructs the AI to act as a wellness coach, not a clinician
• Emergency signposting is surfaced automatically when low mental health scores are detected, regardless of AI output
• Persistent disclaimers appear on every results page
• Users are reminded to review recommendations with a professional

No Automated Decision-Making with Legal Effect

NurtureArc does not use AI to make decisions that have legal or similarly significant effects on users. Your wellness score is informational only.

Feedback

If you believe the AI produced an inappropriate or harmful recommendation, please email hello@nurturearc.polsia.app with details. We take all such reports seriously.