Exploring the new AI Skill update in Microsoft Fabric: Features and limitations

Today, when data is everywhere, businesses are turning to AI to stay ahead of the curve with quick, actionable insights that drive decision-making. But sifting through these vast datasets is quite overwhelming, especially when one isn’t a data expert. Given this pressing challenge, Microsoft Fabric filled the gap by tapping into the power of Generative AI to make data more accessible and understandable for everyone.

One such AI innovation is Copilot in Microsoft Fabric, a powerful tool that has already transformed how data professionals and business users work with data. With Microsoft Fabric AI Copilot, you can generate code, find answers quickly, and get help with a range of data tasks to make your work much more efficient and streamlined.

While Copilot is a valuable asset, there has been a growing demand for an even more intuitive and user-friendly experience that makes it easier for anyone to perform data tasks on Microsoft Fabric. This is where AI Skills make a difference! By offering a more conversational and less hands-on approach to interacting with data, AI Skills, a new update in Microsoft Fabric, can make data more accessible to everyone. By leveraging the power of AI, these skills provide a more intuitive and accessible way to interact with data, allowing users of all levels to gain valuable insights with ease and precision.

Let’s look into this new capability in Microsoft by going through the overview of Microsoft Fabric AI Skill, its key features and benefits, and how it will revolutionize the way businesses extract insights from their data.

A quick overview of what is Microsoft Fabric

Microsoft Fabric is a unified, all-in-one data analytics platform that combines all data, analytics, ML, and AI tools into a cohesive, AI-powered platform. This end-to-end data analytics solution empowers organizations to harness their data more effectively, enabling seamless collaboration, data governance, and insights generation across all departments.

One of the standout features of Microsoft Fabric is its deep integration with generative AI, which significantly enhances the platform’s analytical and decision-making capabilities. Building on these generative AI capabilities, Microsoft Fabric introduces AI Skills, allowing users to interact with their data in a more natural and intuitive way with a conversational interface.

What is AI Skill in Microsoft Fabric?

AI skills integration in Microsoft Fabric is a powerful tool that allows you to create your own conversational Q&A systems using Generative AI. With a bit of setup, you can create a system where your colleagues can ask questions and receive reliable, data-driven answers effortlessly. You can guide the AI system by providing specific examples and instructions, helping it grasp your organization’s unique context and data.

After configuring the AI skill, you can share it with your colleagues. They can then ask their questions in plain English without needing any technical know-how. The AI system takes these questions and converts them into queries that pull the necessary information from your data. It then delivers the answers in an understandable format, making it easy for everyone to get the information they need quickly and efficiently. With AI Skill, Microsoft Fabric developers can now create their own Generative AI experiences using the data stored in their Lakehouse.

But that’s not all. Key benefits of AI Skill in Microsoft Fabric that make an impressive update include:

• Allows users to ask questions in their natural language and receive relevant answers based on the data stored in the Lakehouse.

• Leverages advanced AI models to generate accurate and informative responses.

• Provides valuable insights into your data by answering questions about its content.

• Can be tailored to specific use cases and organizational needs.

• Makes data more accessible to a wider audience by eliminating the need for complex SQL queries.

Are both AI Skill and Copilot in Microsoft Fabric the same?

It’s fair to question if AI Skill and Copilot are somehow related or the same. No, Microsoft Fabric AI Copilot and AI Skill are not the same. Although both are built on Generative AI capabilities, they serve different purposes. Let’s understand the quick difference between the two:

AI Skill

Microsoft Fabric AI Skill is primarily designed to create conversational Q&A systems. Users can ask questions in natural language, and the AI Skill will provide relevant answers based on the data stored in your Lakehouse. Moreover, you can configure the AI Skill system by providing it with a particular set of instructions and examples that tune it the way you need.

Copilot

On the other hand, Copilot is an AI assistant that helps with various tasks, including generating code, exploring data, and automating processes. Copilot is designed to assist with various functions and typically requires more active input and guidance from users. Unlike AI Skill, Microsoft Fabric’s Copilot does not offer configuration flexibility.

Now, another key aspect to understand is how does AI Skill work in Microsoft Fabric?

To help you understand how AI Skills produce responses, we’ve broken down the process in the following steps:

Generative AI and Large Language Models

The AI Skill uses Generative AI, specifically large language models (LLMs), to make interacting with your data easier. LLMs are advanced AI systems trained to understand and generate human-like text based on given inputs.

Generating queries

When you ask a question, the AI Skill sends this question along with details about your data—such as table and column names, and data types—to the LLM. The LLM then generates a T-SQL query designed to answer your question.

Query validation and results

Before executing the query, the system checks it to ensure it won’t alter your data in any unintended way. After the validation, the query is run against your database and display the results to you.

To be precise, AI Skills leverages LLMs to automatically create and execute T-SQL queries based on your natural language questions, streamlining the process of retrieving data insights.

What are the core features of Microsoft Fabric’s AI Skill?

AI Skill empowers users to extract more value out of their data with its compelling array of features that make data interactions more productive and intuitive. Key Microsoft Fabric AI Skill features include:

• Focused AI access

Specify which data the AI should access by selecting particular tables in your databases, ensuring it only interacts with the relevant information.

• Easy instructions configuration

Guide the AI Skill system behavior by providing clear instructions in English, helping it adhere to specific rules or definitions.

• Example-based configuration

Train the AI Skill system by providing example questions and query pairs so the AI can learn and replicate the process for similar queries in the future.

• Strong data security

AI Skill ensures data governance by using the Fabric account of the user (whoever asks questions) to execute queries. Additionally, each query is carefully checked to prevent any accidental deletion or alteration of the data.

Understanding the limitations of Microsoft Fabric AI Skill

AI Skill is currently in public preview. As it is still in its early stages, it has some limitations. There may be instances where it struggles to provide accurate or comprehensive answers. Additionally, the AI Skill may have limitations in understanding complex queries or nuanced requests at this stage. Here are some limitations that you must watch out for:

• The AI Skill can only work with the data you specifically provide and relies solely on the configurations you set up.

• Currently, you can only choose one warehouse or Lakehouse at a time.

• The AI Skill doesn’t have a conversational interface—it can’t remember previous questions, so each question needs to be complete on its own.

• It only accepts questions or instructions in English, blocking any non-English inputs.

• You can’t connect the AI Skill with Fabric Copilots, Microsoft Teams, or other tools outside of Fabric workspace.

• You’re unable to switch the large language model (LLM) that the AI Skill uses.

• If you use non-descriptive column names, the AI Skill might not be as accurate.

• Accuracy can drop when working with large schemas that have many tables.

• Using too many columns and tables might affect the performance of the AI Skill.

• For now, the AI Skill is designed for handling simple queries, rather than complicated ones.

• It’s important to note that AI Skill might give incorrect answers. So, it is good to test it with your colleagues to make sure it’s responding correctly. If you notice any mistakes, you can improve its accuracy by giving it more examples and clearer instructions.

Make smarter, better decisions with your data using Microsoft Fabric

To wrap things up, Microsoft Fabric AI Skill is a game-changer for businesses looking to create their own conversational Q&A systems using generative AI. By allowing users to ask questions in plain language and get accurate, data-driven answers, AI Skills make it easier for everyone to tap into valuable insights without needing deep technical knowledge.

To get started with Microsoft Fabric, benefit from AI Skill in Microsoft Fabric, or for consultancy and implementation of this all-in-one analytics solution, contact us at marketing@confiz.com.