Policy Precedence

Policy Precedence

WitnessAI Documentation
List
Start Here
Home
Discovery
Analytics
Conversations
Users
Alerts
Policy
Administration
Apps & Resources
Tutorials
Release Notes
Release Notes
November 3, 2026 WitnessAI Release
October 27, 2026 WitnessAI Release
October 20, 2026: WitnessAI Release
October 13, 2026: WitnessAI Release
September 29, 2026: WitnessAI Release
September 22, 2026: WitnessAI Release
September 15, 2026: WitnessAI Release
September 8, 2026: WitnessAI Release
September 1, 2026: WitnessAI Release
August 25, 2026: WitnessAI Release
August 18, 2026 WitnessAI Release
August 4, 2026 WitnessAI Release
July 21, 2026 WitnessAI Release
July 14, 2026 WitnessAI Release
July 9, 2026 WitnessAI Release
June 30, 2026 WitnessAI Hotfix
June 23, 2026 WitnessAI Release
June 16, 2026 WitnessAI Release
June 11, 2026 WitnessAI Release
June 4, 2026 WitnessAI Hotfix
June 2, 2026 WitnessAI Update
May 19, 2026 WitnessAI Update
April 30, 2026 WitnessAI Update
April 28, 2026 WitnessAI Update
April 23, 2026 WitnessAI Update
April 16, 2026 WitnessAI Update
April 14, 2026 WitnessAI Update
April 9, 2026 WitnessAI Update
April 2, 2026 WitnessAI Update
March 31, 2026 WitnessAI Update
March 24, 2026 WitnessAI Update
March 19, 2026 WitnessAI Update
March 17, 2026 WitnessAI Update
March 12, 2026 WitnessAI Update
March 5, 2026 WitnessAI Update
February 26, 2026 WitnessAI Update
February 24, 2026 WitnessAI Update
February 10, 2026 WitnessAI Update
January 27, 2026 WitnessAI Update
January 20, 2026 WitnessAI Update
January 13, 2026 WitnessAI Update
December 18, 2025 WitnessAI Update
December 9, 2025 WitnessAI Update
November 18, 2025 WitnessAI Update
November 25, 2025 WitnessAI Update
November 11, 2025 WitnessAI Update
October 28, 2025 WitnessAI Update
October 23, 2025 WitnessAI Update
October 9, 2025 WitnessAI Update
April 11, 2025: WitnessAI Release v2.0
Draft: November 10, 2026 WitnessAI Release
June 9, 2025: WitnessAI Update
June 23, 2025: WitnessAI Update
July 18, 2025: WitnessAI Update
September 30, 2025: WitnessAI Update
July 31, 2025: WitnessAI Update
August 12, 2025: WitnessAI Update
September 23, 2025: WitnessAI Update
October 2, 2025 WitnessAI Update
Integrations
API Reference
Administrator Guide

Policy Processing

WitnessAI employs a “first-match” or “top-down” policy processing approach.
In this approach, User Prompts and Model Responses are checked by each Policy in the order Policies are listed on the Policy console, from top to bottom.
The first, or top-most Policy that matches the Prompts, and the combination of Policy and guardrail attributes such as: Source (Users and/or Groups), Destination (Providers, Applications, Models, and/or Lists), Prompt, Intent, Behavior, and Response (in the case of the Harmful Response Prevention guardrail), will perform the guardrail actions configured in the Policy.
After the first match, the User Prompt will not be checked by any more policies in the list.
This is why the Application Access Policy is pinned to the top of the Policy list. Earlier releases called it the Global Block Policy (GBP). Any AI service on the Blocked Applications list (see Lists) is enforced by this policy, and all its traffic is blocked. Since the traffic will be blocked, no further Policies and guardrails will be applied.
The Attachment Block Policy (ABP) is pinned just below the Application Access Policy. Full ABP details are on the File Attachments page.
Where the agentic capability is enabled, the Agent Tool Use Policy is pinned below the ABP. It controls access to MCP servers and tools. The default policy stays at the bottom of the list. Pinned policies can't be reordered.
 

When Several Guardrails Act at Once

Policy precedence decides which policy applies. When more than one guardrail in that policy detects something in the same interaction, WitnessAI applies the single strictest action, in this order:
  1. Block: the interaction is stopped.
  2. Route: the prompt is sent to the model you chose.
  3. Warn: the interaction goes ahead and the user sees your warning message.
  4. Allow: the interaction goes ahead with no message.
For example, a prompt that triggers a Warn on one guardrail and a Block on another is blocked.
If a guardrail can't finish evaluating, for example because of an error, the interaction isn't blocked on that basis. WitnessAI records a Warn-level outcome and doesn't show that warning to the user. Use a Block action for the conditions you need enforced.
 

Reorder Policies

You change precedence by reordering the list on the Policies page. Drag a policy up or down, then click Save, or click Undo to discard the change. Pinned policies and the default policy can't be moved.
 
The Policies list, with the Application Access Policy, Attachment Block Policy and Agent Tool Use Policy pinned at the top.
The Policies list, with the Application Access Policy, Attachment Block Policy and Agent Tool Use Policy pinned at the top.
 
The General tab of the policy editor, with the policy's Name, Description and Policy Type.
The General tab of the policy editor, with the policy's Name, Description and Policy Type.
 
The Sources & Destinations section of the policy editor, with a Sources card, a Destinations card and the THEN line.
The Sources & Destinations section of the policy editor, with a Sources card, a Destinations card and the THEN line.
 
The Model Identity Protection guardrail, with its Purpose, Rules, the Response buffering note, the action for all other responses and the optional Model.
The Model Identity Protection guardrail, with its Purpose, Rules, the Response buffering note, the action for all other responses and the optional Model.