Responsible AI Statement

MatrioshkaLabs commits to ethical AI development across HushAI, HushSafe, and all intelligent features.

Principles

• Privacy by design — minimize data collection and prefer on-device processing where possible
• Human oversight — appeal flows and human review for edge cases
• Bias mitigation — regular model evaluation and published model cards
• Transparency — clear disclosure when users interact with AI systems

Training & Deployment

Models are trained on curated datasets with safety filters. We do not sell user data for model training. Enterprise customers may configure policy boundaries via HushSafe.