Trust architecture for modern home care
Caire handles sensitive operating data with clear permissions, traceability, data minimization, and controlled workflows. The goal is to make AI and planning audit-friendly for leadership, planners, and care teams.
Current page
Security & compliance
Permissions, traceability, data protection, and controls.
Product trust
Built for responsibility, not black boxes
Caire connects data, recommendations, human decisions and outcomes so responsible users can review what changes and why.
Role-based access
Planners, leaders, caregivers, clients, and family members get different views and permissions based on role and consent.
Audit trails
Decisions, changes, publishing events, and mobile activity can be traced so the organization understands who did what.
Data minimization
Pages and flows show relevant information to the right role without unnecessarily spreading internal operating data.
Human control
AI recommendations remain decision support. Responsible users review, approve, and publish.
Responsible AI
AI compliance belongs in the same trust architecture
AI in Caire is decision support for planning, not autonomous public authority decision-making. Recommendations should be transparent, human-in-the-loop, and reviewable after the fact.
Human-in-the-loop
AI creates candidates, but responsible planners review, adjust, and approve before a schedule is published.
Transparency
Users should understand why a recommendation appears, which goals it affects, and what trade-offs it creates.
Reviewability
Recommendations, changes, and decisions are saved as operational traces for follow-up and internal control.
Risk-aware scope
AI is used for planning and operational support with clear boundaries, responsibilities, and escalation paths.
EU AI Act
Which parts of Caire use AI, how we assess them under the EU AI Act, what can happen automatically and who has the final say.
Data privacy and sovereignty
Customer operating data is handled for the organization’s workflow. Caire does not position public scheduling optimization as hidden generative-model training on customer records.
Constraint governance
Hard rules, Soft goals, clear approval
Hard constraints
Skills, no-overlap rules, time windows, permissions, pinned visits, and safety-critical limits are treated as rules the planning flow must respect.
Soft constraints
Continuity, travel time, workload balance, unused hours, and preference goals are optimized transparently so planners can understand tradeoffs before publishing.
AI governance
AI recommendations are transparent, planner-approved, and scoped to operational decisions. The product is designed so organizations can document controls and responsibilities.
Documentation-ready
The product is designed to help organizations explain data sources, human oversight, evaluation routines, and operational controls when they document their AI use.