Trust Center

Marsin uses tenant-scoped access controls, secure authentication, encrypted transport, encrypted storage for integration credentials, audit records, rate controls, and environment-scoped secrets. Customer administrators control member access, integrations, approvals, and agent autonomy.

Marsin identifies its AI assistant and records the source context, model-generated work, review state, and approval history available to the workflow. Public or consequential marketing actions can be gated for human approval. Higher autonomy is earned through observed reliability rather than enabled by default.

AI-generated images and other assets should be reviewed before publication. Marsin retains provenance fields inside the product and is evaluating additional interoperable machine-readable content credentials as standards and provider support mature.

Customer data is used to provide the requested service and is not sold. Our Privacy Policy, Data Processing Addendum, and Subprocessors page describe roles, providers, transfers, retention, and rights.

We monitor production errors and service health, use managed infrastructure with backup and recovery capabilities, and maintain operational procedures for incidents, failed work, notifications, and customer offboarding. Any contractual uptime or response commitment is stated in the customer's order form.

To report a suspected vulnerability, email security@marsin.ai with steps to reproduce and potential impact. Do not access customer data, degrade the service, or disclose an issue publicly before we have had a reasonable opportunity to investigate. Privacy incidents can be reported to privacy@marsin.ai.

Marsin does not claim a certification or independent assurance report unless it is explicitly published here or provided under confidentiality. Customers may request current security and privacy materials during procurement.