Trust
How Solomon handles your data and its own decisions.
Security, data handling and AI governance, stated as mechanisms you can verify.
Security
- Tenant isolation
- Each tenant’s data lives in its own database schema. The tenant for every request comes from the verified identity token, never from the request itself.
- Authentication
- Single sign-on through your identity provider, with OIDC authorization code and PKCE. Sessions fail closed when a token can’t be refreshed.
- Secrets
- Credentials are held in a managed secrets store and fetched at runtime, never stored in configuration.
- Disclosure
- Report a vulnerability to {{CONFIRM: security contact}}. We acknowledge within {{CONFIRM: n business days}}.
Data handling
- Where data lives
- In Solomon’s hosted service in the United States. Customer-cloud deployment: design-partner preview, on request.
- What Solomon stores
- Normalised events, entity facts with their timestamps, case records and analyst corrections.
- Retention
- Configurable per tenant. Defaults: {{CONFIRM: retention defaults}}
AI governance
- Evidence first
- Agents may only state conclusions they can link to evidence.
- Bounded actions
- A deterministic policy layer between agent reasoning and action execution decides whether an action runs, waits for approval or is only observed. Approval is the default.
- Human in the loop
- Containment and remediation require approval unless you’ve set a policy that allows them.
- Audit
- Every agent action, approval and rejection is logged with who, what and why.
- Your data and models
- {{CONFIRM: whether customer data is used to train shared models}}



