Privacy Policy
Last Updated: March 2026
Guardrail is a service developed by Sentient Index Labs & Technology designed to help organizations evaluate prompts and interactions with AI systems for compliance, governance, and risk indicators before they are processed by external AI models.
Information Processed
Guardrail may process the following information:
- User prompts submitted to the platform
- Metadata associated with requests (timestamp, request size, system status)
- System diagnostic information necessary for service reliability
Guardrail is designed as an evaluation and oversight layer. The platform analyzes prompts to identify potential compliance or risk signals prior to interaction with external AI systems.
Prompt Storage
Depending on deployment configuration:
- Prompts may be temporarily processed in memory for analysis
- Prompts may optionally be logged for audit and debugging purposes
- Organizations may configure their own data retention policies
Sentient Index Labs & Technology does not use prompt content for model training.
Security
Guardrail implements industry-standard security practices including:
- Encrypted network connections (HTTPS)
- Controlled system access
- Server infrastructure monitoring
Third-Party AI Systems
Guardrail may interface with external AI model providers when configured by the user. Those providers operate under their own privacy policies.
Contact
For privacy inquiries: