AI Governance
How AI features work in ClinixQM and our approach to responsible AI in regulated environments.
How AI Works in ClinixQM
ClinixQM uses AI to assist users with document writing, risk analysis, and other quality management tasks. AI features are designed to augment human decision-making, not replace it.
- Document assistance: AI helps draft, review, and improve quality documents based on templates and best practices
- Risk identification: AI suggests potential hazards and risk factors based on product descriptions
- Compliance guidance: AI provides regulatory guidance and requirement mapping suggestions
- Quick helper: AI assistant answers questions about the platform and quality management processes
Architecture & Data Handling
AI processing happens entirely server-side with careful controls around data handling:
- Server-side only: All AI processing happens on our backend servers. No client-side API keys or direct model access.
- Tenant isolation: AI requests are scoped to your organization's data only
- No training on your data: Your organization's content is not used to train AI models
- Logging: AI usage is logged for audit purposes with prompts and responses recorded
Human Oversight
AI in ClinixQM is designed to assist, not automate. Human review and approval is always required:
- Suggestions, not decisions: AI provides suggestions that users can accept, modify, or reject
- Document approval unchanged: AI-assisted documents still require normal approval workflows
- Risk assessments require review: AI-suggested risks must be reviewed and confirmed by qualified personnel
- Clear labeling: AI-generated content is marked as such in the interface
AI Provider
ClinixQM uses OpenAI's API for AI features. OpenAI is listed in our subprocessors. We use OpenAI's enterprise API with data processing agreements that prohibit use of customer data for model training.
Fair Use & Limits
Each plan includes AI token allocation with fair-use monitoring. We monitor for unusual usage patterns but do not impose hard limits that would disrupt normal work. Enterprise customers can discuss higher allocations if needed.
Important: AI suggestions should always be reviewed by qualified personnel. AI is a tool to assist quality professionals, not a replacement for expertise and judgment.