Responsible AI
AI is used where it creates value, connected to trusted knowledge, and governed with permissions, evaluation, and human approval where risk requires it.
Trust center
Controls that belong in the work, not beside it.
Security, quality, and responsible AI are part of how LVC designs and operates systems.
Security in the design
Security is part of architecture: integrations, cloud, data, access, and the operational model. It is not a checklist applied after the build.
Quality in the release
Functional, automated, API, performance, accessibility, and AI testing are selected for the system being shipped.
Audit and evaluation
Production AI is measured for accuracy, groundedness, tool execution, cost, and latency, with a path to review what the system did.
Program constraints
Enterprise and government work proceeds subject to contracting conditions, security requirements, data residency, staffing restrictions, and other applicable program requirements.
Disclosure
Security concerns can be raised through LVC. Use the security disclosure page to see how to send a report.
