Understanding the AI Builder


Course Introduction
Connecting Data with Connectors
Course Summary
1m 14s

The course is part of this learning path

Start course

This course explores the core components of Microsoft's Power Platform, including the Microsoft Dataverse, common data model, compliance, connecting data to Power Platform, and the AI Builder.

Learning Objectives

  • Learn about the Microsoft Dataverse and the Common Data Model
  • Learn about Power Platform environments and data compliance
  • Understand the types of connectors you can use within Power Platform
  • Explore the AI Builder and various low- to no-code use cases for various AI Models

Intended Audience

This course is for anyone who wants to understand the core components within the Power Platform and who wants additional insight into how their data connects within it.


To get the most out of this course you should already have a basic understanding of Power Platform. Before embarking on this course, we recommend that you take our course "Understanding the Business Value of Microsoft Power Platform" beforehand.


The AI Builder is a tool that allows you to add AI into your Power Automate and Power Apps solutions. The reason the AI Builder is such a powerful tool is because it is specifically designed through a unique Point and click experience allowing the use of AI without requiring knowledge of coding. You can utilize one of many prebuilt AI models or custom build them to meet your specific needs. Things like automatically translating support tickets into your language, identifying, and classifying customer feedback on social media, or even automating inventory are all examples of how the AI Builder can help streamline your organization's processes and save you precious time.

The AI Builder prebuilt model types are Category Classification, entity extraction, key phrase extraction, language detection, sentiment analysis, text translation, business card reader, text recognition, and receipt processing. You can also custom make AI for things like object detection, form processing, and even AI prediction. Understanding the capabilities of the AI Builder is important to being able to provide an entire solution to customers utilizing Power Apps and Power Automate, so let's go over some examples.

Let's say Company A has been having an employee manually intake receipts for expenses, input them into Excel, and then saving that data and sending it to management for review and approval. Having every step of that be manual includes five steps: intake of the receipt, collection of the data, adding that data to a larger Excel Spreadsheet, sending that sheet to management, and then having management approve that sheet for payment. That is five different manual steps all of which could be automated using the receipt processing model within Power Automate and Power Apps.

This model takes images of receipts, detects data within them, and outputs the data like Merchant name, cost, dates, and more into a streamlined format. This completely removes the need for manual data entry, and while that is a large time save for many organizations, this is still only a partial solution. You can use this model alongside Power Automate to automate nearly the entire process, so let's think of a flow to showcase this.

As I just mentioned, the five steps in invoice processing for this example are receiving the receipt, collection of the data, adding that data to a larger Excel sheet, sending that sheet to management, and then having management approve that sheet for payment. Now, every flow needs a trigger which can create multiple actions thereafter. And in this example, the trigger would be receiving the receipt. Once the receipt is received, the AI Builder would automatically collect that data and input it into a new data form which would be steps two and three completed.

Then, once the receipt has been processed, we could add an additional action into the flow to send the cataloged data to the manager and request approval thus automating steps four and five. You could even go further and have that approval or denial be sent back to the finance team for final processing. And this is just one example of how AI Builder can help automate tasks within your organization. You can use the Object detection model to detect your organization's inventory through images or detect positive or negative sentiment of your organization on different social media platforms through the sentiment analysis model.

There are so many models and when utilized properly can save your organization large amounts of time. And since AI Builder uses a streamlined Point and click experience the barrier to entry is next to nothing making the majority of the battle simply knowing what tools are available to you rather than the actual creation and utilization in your solution. For more information on all of the AI models, please take a look at the Microsoft documentation on AI Builder.

About the Author
Learning Paths

Lee has spent most of his professional career learning as much as he could about PC hardware and software while working as a PC technician with Microsoft. Once covid hit, he moved into a customer training role with the goal to get as many people prepared for remote work as possible using Microsoft 365. Being both Microsoft 365 certified and a self-proclaimed Microsoft Teams expert, Lee continues to expand his knowledge by working through the wide range of Microsoft certifications.