Security and governance are core to Switchboard. We give organizations centralized control over which AI models employees can use, how data is handled, and how spend is governed, backed by platform-level safeguards.
Encryption in transit and at rest
Traffic is encrypted with TLS in transit. Stored secrets, including provider API keys, are encrypted at rest with AES-256-GCM, and passwords are hashed with argon2.
Access controls
Role-based access, department-level scoping, and two-factor authentication (MFA) with recovery codes. Enterprise SSO via SAML and OIDC is available on Enterprise plans, with Okta, Entra, and Google. SCIM provisioning is on the roadmap.
Data residency awareness
Every model surfaces the region its data is processed in, so you can enforce regional requirements by team.
No training on your content
Switchboard does not use your prompts or content to train models, and surfaces each provider's training practices.
Governance and audit
Model allowlists, provider blocking, budget enforcement, and audit log exports give you a complete control surface.
Reliability
Automatic failover routes around provider outages so your teams stay productive.
Shared responsibility
Switchboard secures the platform; your administrators configure the governance policies that fit your organization: which models are enabled, how budgets are enforced, and how data is retained.
Provider transparency
Because Switchboard sits in front of many providers, we make each model's compliance posture visible (data residency, retention, training practices, and certifications) so your decisions are informed rather than assumed. We state it per model rather than per provider, because a model reached over a hosted route is processed on that host's infrastructure, not its maker's.
Reporting a concern
To report a security concern or request our current compliance documentation, contact hello@theswitchboardapp.com.
Questions? hello@theswitchboardapp.com