We’re excited to share what’s new with SharePoint Premium content services this summer! In this round of updates, we’re sharing details on the general availability of autofill columns and the sensitive information prebuilt model, along with Graph API documentation to process models.
Autofill columns in SharePoint libraries
It’s here – the new SharePoint autofill columns feature for document libraries is now 100% generally available, and it’s a game changer for managing your files and metadata.
Autofill columns use AI to extract, summarize, generate, or evaluate content from the files you upload to a SharePoint library. You just need to write a simple prompt in natural language, and the system will save the answer to the column. You can ask “What is the deadline for this project” or state “Summarize this document in three sentences”, then run the Autofill action, and the column will populate for each file in that library, and all new files added - whether the answer is short text, longform response, numbers, date, choice, hyperlink, or currency.
Here are some examples:
| Prompt | Sample Response |
| [Extract] What is the end date of the agreement? |
August 3, 2024 |
| [Classify] If this is a legal document, respond with “True” or “False” only. If the document doesn’t provide enough information, leave the response blank. |
True |
| [Generate] Summarize this document in 30 words or less |
This document outlines a consulting services agreement between Jon Smith and a company for a home renovation design project, including scope, responsibilities, fees, and completion criteria. |
| [Evaluate] Summarize this document in 30 words or less and apply the most relevant Taylor Swift song title based on the sentiment. |
This document outlines a consulting services agreement between Jon Smith and a company for a home renovation design project, including scope, responsibilities, fees, and completion criteria. The sentiment of the document is professional and forward-looking. The most relevant Taylor Swift song title that reflects the sentiment of this document could be “Begin Again”, as it suggests a new start or project, which aligns with the nature of a consulting agreement for home renovation. |
|
[Impersonate] Imagine you’re a pirate, respond with a short summary like a pirate would write. |
Arrr, this document be a contract with Jon Smith and a company to renovate his ship, matey. |
Autofill columns work with common file types such as PDF, Word, Excel, PowerPoint, etc. They can also work in conjunction with other SharePoint Premium models, such as prebuilt models, to extract values that the other model may not get.
Getting your autofill columns set up is simple. Once your admin enables it, you’ll add a new column in your SharePoint library, and select the autofill option. Then, write your prompt or use one of the suggested prompts provided. Test your prompt on any file in the library and see the result. When you're happy with your prompt, save it and apply it to files you select and choose the “Autofill” action.
Autofill columns are rolling out now, and we hope they’ll help you find, sort, filter, and analyze your files more easily and efficiently. Learn more: Overview of autofill columns Microsoft Syntex - Microsoft Syntex | Microsoft Learn
Sensitive information prebuilt models
Document processing models help you automate the classification and extraction of information from documents – such as forms, invoices, contracts, and receipts – with either custom models that you build, or prebuilt models that we provide. Today, we’re excited to share that a new type of prebuilt model is now 100% generally available, to help those of you who work with documents containing sensitive information, such as personal or financial data.
With the sensitive information prebuilt model, you can swiftly detect and optionally extract email addresses, phone numbers, and physical addresses from countless documents. In just a few clicks, organize this vital information in your SharePoint library.
No coding or training is needed to use this sensitive information prebuilt model, and you can configure it in a few simple steps. Choose the language and the entities you want to detect and / or extract, and then test the model on some sample files. Once you're happy with the results, apply the model to any library. The model will then display the detected entities in a column, and if you chose to extract values, those are put in their own columns. You can use these columns to sort, filter, or analyze your files. The sensitive information prebuilt model can even read printed or handwritten text. In the future, we intend to support this model for detection/extraction on encrypted files, and apply sensitivity or retention labels just as you can with other models today.
Page range support
Speaking of processing models, you can now specify the page range to extract information from when you’re using structured, freeform, and prebuilt document processing models. Now you don’t have to extract from an entire file, and can process just the pages you select. This can be a great cost and time savings by reducing the number of pages processed.
Monthly included capacity promo – now extended through June 2025!
You’ll recall that we’ve been running a promotional offer that gives you monthly included capacity for many SharePoint Premium pay-as-you-go services. This offer is now extended through June 2025!
Both autofill columns and the sensitive information prebuilt models are part of the monthly included capacity promotional offer. Autofill columns are under the unstructured documents processing service, and with this promo, you can process up to 100 pages per month with autofill at no cost. Similarly, for the prebuilt models for sensitive information you’ll be able to process up to 100 pages each month at no cost.
We hope you’ll give both autofill columns and sensitive information prebuilt models a try. Let us know in the comments how you’re using these AI-powered content management capabilities.
SharePoint Content AI Model Graph APIs
For customers and partners alike, we’re announcing the launch of SharePoint Content AI Model Graph APIs. These models, which include unstructured, structured, freeform and prebuilt document processing, are now accessible through Graph APIs. These APIs enable developers to list existing models on a library, apply or un-apply a model, and create processing jobs for document metadata extraction from, and labeling of, your content. The launch introduces two new resource types to Microsoft Graph: 'content model' and 'document processing job.' These additions come with new methods that allow developers to:
| Get | Read the properties and relationships of a contentModel object. |
| Get applied drives | List all the contentModelUsage information related to an applied contentModel. |
| Get by name | Read the properties and relationships of a contentModel object by its model name. |
| Add to drive | Automatically process new documents in the library. |
| Remove from drive | Remove the contentModel from the document library. |
| Create | Create a new documentProcessingJob object. |
Developers can now build sophisticated applications that integrate with SharePoint's existing Graph APIs. For instance:
• An application for document metadata extraction can work with Exchange and monitor an email inbox, extract email attachments to SharePoint, and use content AI models to extract critical metadata.
• A content migration application can upload documents from a network drive to SharePoint Online, using content AI models to set content type, extract metadata, and apply sensitivity and retention labels.
These new Graph APIs offer unified authorization and SDK experiences, so it’s easier for developers to build more complex apps. Additionally, developers can programmatically manipulate content AI models and process files with or without UI interaction, streamlining the development process. Plus, leverage Graph calls to harness the full potential of Content AI with SharePoint Embedded! Learn more in our Graph API documentation for contentModel and documentProcessingJob.
Content Management Advocates spotlight
Earlier this year, we launched a new initiative to share the great work of the many thought leaders in the content management space. Check out the work of some Content Management Advocates below!
• Read Leon Armston’s timely new blog post on Autofill Columns. Learn how to enable it in your tenant for users, get more prompt ideas, and combine this with other SharePoint Premium document processing models for even better experiences.
• Watch Steve Corey's video on SharePoint Document Translation which features a tutorial on how to enable and configure the translation feature.
• Learn from Juan Carlos Gonzalez's video on how to set up Pay-as you-go services in SharePoint Premium.
Try out these capabilities today
Since we’ve extended the promotional offer for free monthly capacity of many SharePoint Premium pay-as-you-go capabilities through June 2025, there’s no better time to get started with SharePoint autofill columns and sensitive information prebuilt models. For our developers in the community, give Graph APIs a try and let us know the types of apps you’re building.
Unsure about how pay-as-you-go pricing works or how to manage costs? The SharePoint Premium pay-as-you-go calculator is a helpful tool for predicting costs associated with using our content management capabilities. This calculator gives you a better understanding of your organization’s usage patterns and estimated costs so you can make more informed decisions with your budget.
Microsoft
