---
sourceDocument: Xanadu Enable AI
sourceDocumentLink: https://servicenow-prod.fluidtopics.net/r/xanadu/intelligent-experiences

 Release :

    - xanadu

ft:locale :

    - en-US

ft:publication_title :

    - Xanadu Enable AI

ft:clusterId :

    - platai

bundleId :

    - platai

workflow :

    - Platform


---

# Data extraction modes in Now Assist in Document Intelligence

# Data extraction modes in Now Assist in Document Intelligence {#ariaid-title1}

* Release version: Xanadu
* 
* Updated May 31, 2025
* 
* ![](https://www.servicenow.com/docs/portal-asset/ico-clock) 1 minute to read

The extraction mode determines how Now Assist in Document Intelligence processes a document task. Extraction modes only apply to use cases for the document extraction skill.
Now Assist in Document Intelligence uses the following data extraction modes.{#data-extraction-modes-now-assist-document-intelligence__table_iqg_n4k_w2c__entry__2}

| Extraction mode | Description |
|-|-|
| Default | Now Assist auto-fills each field in the Document Intelligence workspace if it recognizes a value for it in the document. Now Assist may not always produce accurate, complete, or appropriate information. All fields should be reviewed for accuracy. A reviewer can accept the predicted value or manually enter a value. |
| Full automation | Now Assist automatically completes and submits the document task without an agent review. The values for all required fields are auto-filled by Now Assist or are identified as missing in the document. By turning on Full automation mode, the agent review used to check the accuracy of the predicted values is bypassed. There is a configuration option to suspend the automation and require an agent review when any of the required fields are missing in the document. |
[ ]

{#data-extraction-modes-now-assist-document-intelligence__table_iqg_n4k_w2c}
**Related reference**   

* [Document and visual insights AI agent](https://servicenow-prod.fluidtopics.net/Z_t8H0dCMgpIDljx1t93yg "The document and visual insights AI agent gathers context from user input and image attachments, generates the requested information based on the image content, and provides the information along with any relevant task details.")