CareAI AI Transparency Notice (English)

Effective date: 2026-01-22
Version: 1.1
Last Updated: January 22, 2026

Governing Language: In case of discrepancy between translations, the English version prevails.

CareAI may use automated and AI-assisted functionality to support user experience and provide suggestions. This notice explains how those features work at a high level.

1. What AI may be used for

Depending on enabled modules, AI-assisted features may help with:

Generating reminder phrasing or supportive messages
Summarizing non-sensitive app activity for user-friendly insights
Chat assistance: Providing informational responses to user questions about medications, health topics, and general inquiries
Distress signal detection: Analyzing chat conversations to identify potential distress signals or safety concerns (best-effort basis only)
Content summarization: Summarizing medication information from public databases (e.g., OpenFDA) for informational purposes

2. What AI is not

AI outputs are not medical advice. All AI-generated content is informational only and should not be used to make medical decisions.
AI does not replace a qualified clinician. AI cannot diagnose, treat, or prescribe medications.
AI may be inaccurate or incomplete. AI responses may contain errors, outdated information, or may not be appropriate for your specific situation.
AI does not provide emergency services. AI-assisted features are not a substitute for calling emergency services or seeking immediate professional medical care.
AI does not guarantee detection accuracy. Distress signal detection is provided on a best-effort basis and may fail to detect all signals or may produce false positives.

3. Data used for AI

AI features may use limited user-provided context necessary for the requested function, including:

Chat messages and conversation history (for chat assistance and distress detection)
Medication queries and inquiries (for informational responses)
User profile information (for personalization, where applicable)

We aim to minimize data sent to any AI component and process data in accordance with our Privacy Policy.

4. Health-related AI features

4.1 Medication Information

AI may retrieve and summarize medication information from public databases (e.g., OpenFDA).
This information is informational only and should not be used for dosing, prescribing, or diagnostic purposes.
AI does not provide personalized dosing recommendations or treatment advice.

4.2 Chat Assistance

AI provides informational responses based on publicly available data and general knowledge.
Chat responses are not medical advice and should not replace consultation with qualified healthcare professionals.
AI responses may not be accurate, complete, or appropriate for your specific medical situation.

4.3 Distress Signal Detection

AI may analyze chat conversations to identify potential distress signals or safety concerns.
Detection is not guaranteed: We do not guarantee that all distress signals will be detected.
Accuracy is not guaranteed: Detected signals may be false positives or false negatives.
Not a substitute for professional assessment: AI detection is not a substitute for professional mental health assessment or emergency services.

5. User controls

Where applicable:

Users can choose whether to use AI-enabled features.
Users can disable AI features in Settings if available.
Users can review and correct profile details that influence personalization.
Users can choose not to use chat features if they prefer not to interact with AI.

6. Safety and quality measures

We use safeguards intended to:

Reduce harmful outputs
Prevent unauthorized access
Limit processing to the intended purpose
Filter inappropriate or unsafe content where possible

However, no system is perfect, and AI outputs may still contain errors or inappropriate content.

7. Third-party AI services

CareAI may use third-party AI services to power certain features. These services are subject to their own terms and privacy policies. We select service providers that align with our data protection standards, but we are not responsible for the practices of third-party AI providers.

8. Limitations and disclaimers

No autonomous decision-making: AI does not make autonomous decisions on your behalf. You remain responsible for all decisions.
No replacement of professionals: AI does not replace qualified healthcare professionals, emergency services, or professional medical judgment.
Best-effort basis: AI features are provided on a "best effort" basis. We do not guarantee accuracy, completeness, or availability.

9. Contact

Questions about AI features: support@careai.app

Change Log:

v1.1 (January 22, 2026): Added explicit coverage for CHAT-11 features including chat assistance, distress signal detection, medication information, and health-related AI limitations. Clarified that AI outputs are not medical advice and do not replace professional judgment.
v1.0 (December 17, 2025): Initial version.