In the AI-Powered Prefill feature, once an inbound or outbound call ends, the conversation transcript of that call is sent to the AI, which fills in the necessary data in the fields for the After Call Work (ACW). The agent’s manual intervention is only needed to review and submit the ACW.

This feature enhances the agent productivity by reducing the time needed to complete an ACW, by filling the ACW for the agent, allowing them to handle more calls within the same time frame. This feature also increases the accuracy and efficiency of the data.

When ACW values are prefilled using AI, Sprinklr clearly indicates that the content is AI‑generated and requires agent review before submission.

​

This article is intended for:

• Contact center administrators configuring ACWs
• Supervisors and Customer Experience Managers improving agent productivity
• Teams managing AI-powered automations

At runtime, this feature:

• Sends the call transcript to AI after the interaction ends
• Automatically fills ACW fields with relevant information
• Allows agents to review, edit, and submit the form before completion

​

Prerequisite

Ensure that the Agent Copilot is enabled in your environment to use the ACW Prefill feature. For more details, see Getting Started with Agent Copilot.

Enable AI Autofill for an ACW

To enable AI Prefill:

1. Open your ACW configuration. For information on accessing ACW, refer to the Configuring an ACW article.

2. Toggle Enable AI Prefill Support.

   

3. Add an ACW Description (required).

4. Save the configuration.

ACW Description Requirement

When AI Prefill is enabled, you must provide a short description. While configuring ACW, the information provided in the Description field will be used by the AI model and the AI model leverages the provided context along with transcripts of the case to predict the right value for the field with higher accuracy when using AI to prefill ACW.

• Length: 2–3 lines
• Purpose: Help AI understand the business context

Example:

Retail banking inbound support for card and loan queries. Agents capture product type, issue, and resolution.

​​

Configure Field-Level AI Autofill

AI Prefill Instructions (Required)

Each input field includes AI Prefill Instructions when AI Prefill is enabled.

• Required for all fields without default values

• Provide instructions on:

  • What the field represents
  • How it should be filled

  • Any important business rules

​

​

AI Prefill Example Field

For Picklist and Radio Groups fields, the AI Autofill Examples field is available. You can use this option to add Examples and improve AI accuracy.

• Provide realistic scenarios

• Helps AI learn patterns (few-shot guidance)

Example: Callback Needed? (Picklist)

• Description: Indicate whether the customer has requested a callback.

• Examples per option:

  • Yes: “Call me back later”, “Please schedule a callback”

  • No: Issue handled in the current interaction

​

For more information on configuring Fields, refer to the Adding Fields article.

​

The description field is available for the following Screen components and Disposition Plan fields.

• Text

• TextArea

• Date

• DateTime

• Picklist

• Number

• Phone

• Toggle

• URL

• Email

• Password

• Radio Group

• Custom Field

• Checkbox

• Checkbox Group

Configure Disposition AI Autofill

Refer to Create and Configure a Disposition Plan for more details on configuring a Disposition Plan.

The AI Prefill Settings tab in the Disposition Plan Settings page will be visible only when the Enable AI Prefill Support toggle has been turned on during ACW configuration.

​

​

You can configure AI behavior for:

• Disposition
• Sub-disposition

To configure:

1. Open your Disposition Plan.
2. Navigate to AI Prefill Settings tab.
3. Turn on Enable AI Prefill toggle.

4. Add:

   • Disposition Plan Prefill Instructions
   • Sub-Disposition Prefill Instructions
   • Disposition Prefill Examples
   • Sub-Disposition Prefill Examples

Adding instructions improves prefill accuracy.

​

Examples

​

System Warning for Missing Configuration

If disposition instructions are not configured, a warning appears: Prefill accuracy may be lower without configuration.

​

​

​

​

​

How AI Autofill Works (Agent Experience)

Post-Call Flow

When a call ends:

• The ACW form opens automatically for the agent

• The system sends the conversation transcript to AI

• AI generates values for ACW fields

• The prefilled ACW is presented to the agent for review

What Gets Prefilled

AI Autofill can populate:

• Disposition - The AI assigns a relevant call disposition based on the context of the conversation.
• Sub-disposition - Specific details and outcomes of the conversation are further categorized into sub-dispositions.
• Custom ACW form fields - Any required form data, such as the type of request, product information, or customer feedback, is pre-populated.
• Other input components relevant to the workflow - Other custom ACW screen fields can also be pre-filled based on the specific needs of the business or campaign.

Example Use Case

Imagine a telecom support agent finishes a call where the customer reports slow internet. Sprinklr AI+ analyzes the conversation and automatically fills out the ACW with:

• Disposition: "Network Support"

• Sub-Disposition: "Connectivity Issue"

• Form Field: Customer ID, Issue Description, Resolution (network reset performed)

The agent reviews the details, makes minor edits if needed, and submits the ACW quickly.

​

​

This image illustrates the message that states that the Sprinklr AI+ is filling out your ACW...

Once the data is received from the AI, the fields are automatically pre-filled. The agent can click Next to proceed to next screen or review the form.

Use of Conversation Transcripts for AI‑Powered ACW Prefill

AI‑powered ACW Prefill can use either all conversations in a case or only the latest one.

All Conversations

Use this when you want prefill values based on the complete conversation history (for example, follow‑ups or callbacks).

Add this tag in the Logic Pipeline Prompt: CONVERSATION: {{ADDITIONAL_DATA.UNIVERSAL_CASE.LATEST_VOICE_CASE_CONVERSATION_CONTENT}}

Latest Conversation Only

Use this when you want prefill values based only on the most recent interaction (voice cases only).

• Ensures output reflects the latest interaction
• Prevents outdated information from older conversations
• If there is only one conversation, it is used by default
• If no transcript is available, the system handles it gracefully

​

​

Advantages of using AI-powered ACW Autofill

1. The agents will save a lot of time that can be utilised to take additional calls.

2. The efficiency and the accuracy of the data increases.

For more information on configuring AI-powered ACW Autofill, refer to Configuring AI-powered ACW Autofill.