Create Your Customized Copilot Using Microsoft Copilot Part 2

Introduction

This blog is a continuation of “Create Your Customized Copilot Using Microsoft Copilot Part 1,” in which we explained left side navigation menu Overview, Topics, Actions (preview), Entities, Analytics, Publish, an option to Extend Microsoft Copilot (preview), Settings, that part we will cover in this blog.

Overview of Settings

Unser Settings menu's first tab is "Copilot details" and has two sub-tabs: "General" and "Advanced".

Under "General", there are fields for the copilot's name, which is filled with "Nop Commerce Support", using this field, you can change the name of your copilot. and an option to view the solution, which in this case is "Common Data Services Default Solution" which will open in Power App.

There is also a section for the Icon, suggesting the user can change the icon representing the copilot in chat controls. Current icon, which is a common representation of customer support or technical service, there is a note that the icon should be in PNG format and less than 30 KB in size.

The “AI Integration Tools” under “Settings”, these tools are part of the Microsoft Conversational AI stack which can be integrated with Azure AI Studio and Azure Services.

There are three main integration tools featured in this interface.

• Azure Cognitive Service for Language: This tool allows users to integrate a conversational language understanding (CLU) model with their copilot. Businesses can use this feature to enhance the natural language understanding capabilities of their bots, making them better at understanding user queries. This could be particularly beneficial for customer support or interactive bots that need to decipher various user inputs.
• Copilot SDK: With the Software Development Kit (SDK) for Copilot, developers can extend the capabilities of their Copilot using professional coding practices. This means they can add custom functionalities that are not provided out of the box. It also simplifies the migration to Microsoft Copilot Studio through Skills, which could be useful for businesses that are transitioning from another platform or upgrading their existing systems.
• AI models from Azure Open AI: This feature allows businesses to use Azure Open AI to analyze their own data and ground their AI model in it. By doing so, the AI model becomes more tailored to the business’s specific needs and can provide more accurate generative answers. This is extremely valuable for companies that want to implement AI solutions that require an understanding of specific or technical content, such as legal documents, technical manuals, or customer data.

For businesses, these integration tools mean that they can create more intelligent and responsive bots that can understand and interact with customers in a more natural and helpful way. This can lead to improved customer satisfaction, increased efficiency in handling queries, and a richer, more personalized user experience. Additionally, the ability to train AI models on their own data allows businesses to maintain relevance and accuracy in responses, providing a competitive edge in customer service and support.

Tody everyone is talking about GPT models and businesses are interested in using these models in their businesses to increase efficiency and same you can do this using "Generative AI," under “Settings” which is described as a premium feature within a software platform, part of Microsoft's suite given the mention of SharePoint.

The section includes three primary configuration options.

• Public websites & SharePoint sources: Users can add URLs of public websites or SharePoint sources to the AI model. The purpose here is likely to allow the AI to use the content from these sources to generate responses. There is a field for entering a website URL and an "Add" button, with a list showing already added sources, indicating it's possible to include multiple sources for the AI to reference.
• Upload a document: This feature enables users to upload a text-based document that the AI can use to generate real-time responses. The uploaded document is stored securely and serves as a data source for automatic responses. It’s noted that images, audio, video, and executable files are not supported. There is also an indication that the file contents are available to all users, suggesting that access to uploaded content is not restricted by file permissions or access controls.
• Copilot content moderation: There's a drop-down menu for content moderation settings, with "High (default)" selected. This implies that the copilot's generative responses are moderated for content, and the high setting may result in fewer answers being generated but with more careful content consideration. Lower moderation settings may produce more responses with potentially less content filtering.

Businesses can use these features to provide their virtual assistant with a rich knowledge base to draw from, ensuring more accurate and contextually relevant responses to user inquiries. For example, a company could provide the AI with its website and product documentation to enable the AI to offer detailed product support. Additionally, the content moderation feature helps maintain the quality and appropriateness of responses, which is critical in maintaining professional standards in customer interactions.

Conclusion

Microsoft Copilot is an innovative platform for enhancing customer service with AI, offering tools for language understanding and customizable responses. It enables businesses to create responsive and intelligent virtual assistants that can interact with customers across multiple channels. By integrating with services like Azure Cognitive Services and Azure Open AI, Copilot tailors interactions to be relevant and precise. Generative AI features and content moderation ensure interactions are efficient and professional. Microsoft Copilot equips businesses with the tools to deliver superior customer service and stay ahead in the digital landscape.