---
sourceDocument: Australia Enable AI
sourceDocumentLink: https://www.servicenow.com/docs/r/australia/intelligent-experiences

 Release :

    - australia

ft:locale :

    - en-US

ft:publication_title :

    - Australia Enable AI

ft:clusterId :

    - platai

bundleId :

    - platai

workflow :

    - Platform


---

# Change and offboarding requests

# Change and offboarding requests {#ariaid-title1}

Release version: Australia  
Updated April 23, 2026  
![](https://www.servicenow.com/docs/portal-asset/ico-clock) 1 minute to read  
Make controlled changes to a managed AI asset, or retire an asset that is no
longer needed, by submitting a request for AI steward review and approval.

A request is a structured submission that an asset owner makes when they need to
change or retire a managed AI asset. AI Control Tower routes each request to an AI
steward for review, runs the request through the appropriate approval playbook, and
either applies the requested change or moves the asset to retirement. Requests give
asset owners a controlled way to evolve their AI assets, and they give AI stewards a
consistent way to govern those changes.

## Types of requests {#ac-change-and-offboarding-requests__types}

AI Control Tower supports two types of requests:

Change request
:   A request to modify a managed AI asset. A change request might propose
    switching the AI model that backs an agent, updating an asset's version,
    modifying the asset's use and purpose, or changing the sub-AI systems and
    AI models associated with the asset. Change requests give asset owners a
    path to keep their assets current without bypassing governance.

Offboarding request
:   A request to retire a managed AI asset. An offboarding request initiates
    the offboarding lifecycle stage for the asset, which removes the asset
    from active use and confirms that dependencies have been resolved and any
    data the asset produced is handled according to policy.

## Request lifecycle {#ac-change-and-offboarding-requests__lifecycle}

Both request types follow a similar lifecycle:

1. An asset owner submits the request from the asset record.
2. AI Control Tower runs the request through an approval playbook. The playbook might require a review of the asset, an evaluation of the impact, and a remediation plan before the request can be approved.
3. An AI steward reviews the request, including the impact on related sub-AI systems and AI models. The steward approves or rejects the request.
4. If approved, AI Control Tower applies the change to the asset (for a change request) or moves the asset into the offboarding stage (for an offboarding request). If rejected, the asset is unchanged and the request is closed.

## Where requests appear {#ac-change-and-offboarding-requests__where-they-appear}

Requests appear on the Requests sub-tab of the
Team, Assigned to you, and
Unassigned views in Activity Center, depending on
assignment. Asset owners can also see requests they have submitted from the
affected asset's record.