Settings is where a tenant is configured: who administers it, how AskRAI classifies response confidence, what it does about low-confidence answers, which models it may call, and how external applications authenticate. Open Settings with the Settings control in the top navigation bar — this switches the console to the settings plane, where each area below appears in the left Settings sidebar. Select Console or Projects in the same top bar to leave it.
Settings is Platform Admin only. A Member does not see the Settings control at all, and a Settings link they open sends them back to Projects. Grant the role on the Access page.
Configuration
Users
Manage user accounts, native groups, resource-access roles, and review the synced directory.
Access
Grant the Platform Admin and Member tenant roles to directory users or groups — who may open Settings.
Confidence Thresholds
Tune the score boundaries that classify AI responses into high, medium, and low confidence bands.
Escalation Rules
Define automated actions — tickets, reviews, or suppression — that fire based on confidence scores.
Model Providers
Connect your own AI model providers (BYOM) for use in routing configuration.
Routing
Decide which model answers which query, with rules and an automatic fallback.
Programmable Access
Issue API, CLI, and MCP credentials for external applications and AI agents, with tenant-wide rate limits and per-credential overrides.
Directory Sync
Provision users and groups from your identity provider over SCIM 2.0, or pull them from Google Workspace.
Platform
Alert Preferences
The Alerts section is coming soon. Once available, it will allow you to configure notification preferences for escalation events and system alerts.