Photo by Eugene Chystiakov on Unsplash
Harnessing AI in Microsoft 365: A Beginner's Guide to Azure OpenAI and Microsoft Copilot
The Azure OpenAI Service together with Microsoft Copilot forms an incredibly powerful AI-powered ecosystem and makes tasks easier across applications like Microsoft Word, Excel, and Teams.
My intention with this guide is to help you come away with an understanding of how these things work together and give you clear, step-by-step instructions to enable you to explore them on your own.
You will be able to configure Azure OpenAI in your Azure account at the end, and by then you will understand how it powers Microsoft Copilot's helpful features.
Introduction to Azure OpenAI and Microsoft Copilot
Azure OpenAI Service provides powerful AI models (like GPT-3) via Azure. We can use them for tasks like text generation, summarization, and language understanding. Microsoft Copilot is the built-in AI assistant in Microsoft 365 apps. It uses Azure OpenAI's capabilities to help us automate and simplify tasks. For example, Copilot can summarize Word documents, analyze Excel data, and suggest follow-ups in Teams. Together, they transform how we work and make complex tasks simpler.
Setting Up Your Azure OpenAI Environment
To get started, you’ll need to set up Azure OpenAI in your Azure account. Don’t worry; I’ll guide you through each step.
Step 1: Creating an Azure Account
If you don't have an account there already, you can create one at azure.microsoft.com.
New users get some free credits that can be used to try a whole range of services, including OpenAI.
Once you have created an account, log in to the Azure portal.
Step 2: Navigating the Azure Portal
When you log in to the Azure portal, you’ll see a dashboard with different sections. Please explore the interface.
In particular, check the "Marketplace" and "Resource Groups" sections. You'll need them to create new resources.
You will find both these menus from the left side panel.
Provisioning the Azure OpenAI Service
Now that you’re set up in the portal, let’s create your first Azure OpenAI resource.
Step 1: Creating an Azure OpenAI Resource
- In the Azure portal, search for OpenAI in the Marketplace (you’ll find it in the left-hand sidebar).
- Select Azure OpenAI and click Create. You’ll need to fill out some basic information, like the name of the resource, your subscription, and your region.
- Once you’ve completed these details, click Review + Create and then Create. Azure will now provision your OpenAI resource.
Step 2: Accessing the OpenAI Playground
- After your OpenAI resource is ready, go to it by selecting OpenAI Studio.
- Here, scroll down the page to find the Playground option, a hands-on environment where you can try out OpenAI models with custom prompts.
- In the Playground, you can type in prompts and see the AI’s responses. Try typing simple commands like “Tell me a fun fact” or “Summarize this text: [your text here].” The Playground is a great way to understand how OpenAI’s models respond to different queries.
Understanding the Role of Microsoft Copilot in the Ecosystem
Let’s now look at Microsoft Copilot and how it uses Azure OpenAI to support tasks in applications like Word, Excel, and Teams. Microsoft Copilot acts as a helpful assistant across the Microsoft 365 suite. Using Azure OpenAI, Copilot can understand natural language commands and assist you with a range of tasks.
Use Cases in the Microsoft Copilot Ecosystem
Now, let’s look at how Azure OpenAI’s power behind Microsoft Copilot shines through in practical use cases.
Use Case 1: Document Analysis in Word
- In Microsoft Word, open any document and ask Copilot any question you want.” Let’s open a word file and ask “who is the target audience?” It will analyse the text and provide relevant answers.
Use Case 2: Data Analysis in Excel
- In Excel, select a data set and type in a natural language command like “convert this data to a range of a table.”
- Copilot will create an attractive table based on the data. You can also use suggested options.
Use Case 3: Task Automation in Microsoft Teams
- In a Teams chat, you can ask Copilot to summarise a conversation or suggest responses to messages. This is especially useful for catching up on meetings or managing team updates. See the example in the image below:
Conclusion and Future Exploration
Congratulations! You’ve now set up Azure OpenAI, explored the OpenAI Playground, and understood how Copilot uses these AI models in Microsoft 365 applications. Feel free to experiment further by creating your own prompts, testing different OpenAI models, or building custom applications.
For further learning, check out the official Azure OpenAI documentation and tutorials on Microsoft Learn. Each step you took here is just the beginning of what’s possible with AI and the Microsoft Copilot ecosystem.
Follow Umesh Pandit