---
title: "Build a treatment plan with AI"
canonical: "https://support.appfire.com/space/RM/3459219901/Build%20a%20treatment%20plan%20with%20AI"
format: markdown
---
Appfire LiveRisk uses Forge LLM to analyze risks and generate treatment plans. Similar to [scoring a risk using AI](https://appfire.atlassian.net/wiki/spaces/RM/pages/3459121540), the model can propose a treatment plan and mitigation tasks.

Open a risk work item and go to the **Risk analysis** panel. If the panel isn't visible, select **View app actions** below the work item summary, then select **Risk analysis**.

## Before you begin

- To use AI-assisted treatment-plan actions, you need **Manage LiveRisk Risk Register** and the Jira permissions required to view and edit the Risk work item.
- The **LiveRisk Manager** and **LiveRisk Analyst** roles have **Manage LiveRisk Risk Register** by default.
- The risk must have an **Inherent score** before you can generate a treatment plan.

## Generate a treatment plan

In **Risk analysis**, locate the **Mitigation progress **panel.

1. Click **AI treatment plan**, or if a treatment plan has already been generated, click **Re-run treatment plan**.
2. Review the **AI treatment plan draft**. The plan includes a Current-to-Target score journey, proposed mitigation tasks with score changes, a suggested due date, and the model's reasoning.
3. Use **Refine the plan** to provide additional instructions, or **Re-run treatment plan** to start over.
4. Click **Accept** to open *Confirm treatment plan*.

You can use the thumbs-up or thumbs-down control to rate the usefulness of the AI-generated plan.

## Confirm treatment plan

Risk owners propose the approval modes that determine how Current score and treatment-plan changes are handled. The proposed modes must be accepted through the governance workflow before they take effect. 

When you click **Accept** for a generated plan, the **Confirm treatment plan** panel opens. Configure the following:

![Appfire LiveRisk, confirming the treatment plan.](media://1ebc8dd0-dfd2-42cc-affe-1cf19794f622)

1. **Current score approval mode**:
  - **Manual: AI proposes score changes for acceptance** — you accept each proposed change.
  - **Allow AI to Auto-update score from signals** — accepted signal changes update Current score automatically.
  - **Auto-update score from RTP progress** — treatment progress moves the score (not available on Dynamic risks).

> ℹ️ **RTP approval mode** – Available for dynamic risks:
> ℹ️ 
> ℹ️ - **Manual: AI proposes RTP changes for acceptance** — Proposed changes to the Risk Treatment Plan require acceptance.
> ℹ️ - **Allow AI to auto-create RTP work items from new Signals** — selecting this grants your execution mandate (active only after GRC acceptance).

  - **Due date**
  - **Treatment decision** — **Reduce**, **Retain**, **Avoid**, or **Transfer**
  - **Create tasks in project – **Select the** Jira space** where the mitigation tasks will be created.
  - **Task type** for the mitigation tasks – **Risk**, **Task**, **Risk Treatment Plan**.

The risk owner proposes the approval modes. They must be accepted through the governance workflow before the corresponding automation takes effect. The risk manager can accept the proposed mode or select a less autonomous mode, but cannot grant more autonomy than the risk owner proposed.

> ℹ️ For more information, see **[Risk Manager next action card](https://appfire.atlassian.net/wiki/spaces/RM/pages/3459449335/Risk+manager+next+action+card?atl_f=PAGETREE)**.

## Create the plan and tasks

When you've reviewed the configuration, select **Create plan & tasks**.

LiveRisk:

- creates the Risk Treatment Plan (RTP) work item,
- creates and links the proposed mitigation tasks,
- sets the **Target score**, and
- records the proposed approval modes.

When plan creation runs in the background, **Creating plan and tasks…** appears while LiveRisk creates the RTP work item and its tasks. The panel updates automatically when the process finishes.