Generative AI with Microsoft Fabric

Video script:

– If you want to deliver personalized Gen AI experiences to your app, allowing users to interact with you in natural language, the better the quality and freshness of the data you use to base your responses, the more relevant and accurate the results you generate will be.

– Of course, the challenge is that your data may be spread across multiple clouds, your own data centers, and the edge. This is where Microsoft Fabric, a complete analytics platform, helps you unify data wherever it lives at unlimited scale, without having to move the data.

-Integrate OneLake, a logical multi-cloud data lake for unified data storage and access, and provide a separate real-time hub optimized for event-driven streaming data, where change data capture feeds can be streamed from multiple cloud sources for real-time analytics without the need to import data. Once the data is unified, data experts can work together in a collaborative workspace to collect, transform, analyze, and approve data as they build quality data sets.

– And when combined with Microsoft Purview, this can be achieved with an additional layer of security that allows everyone from engineers, data analysts to business users to classify and protect schematized data as it flows through the Fabric workspace. Keeping LLM’s underlying data up to date is also made easier with the ability to take action with Real Time Intelligence.

-For example, if you have a product recommendation engine on an e-commerce site, you can use real-time intelligence to create granular conditions that listen for data changes, such as new inventory coming in, and update the data pipeline that provides the underlying data for a large-scale language model.

– So while gen AI previously didn’t have access to up-to-date inventory data for ground response, with Real Time Intelligence, the generated response benefits from the most up-to-date, real-time information, so you don’t lose a sale. And when you work with data, the gen AI experience is infused throughout the Fabric. In fact, the Copilot in Fabric experience is available across all Microsoft Fabric workloads to help you work with it.

– Once your dataset is complete, you can easily connect to a large-scale language model in your Gen AI app from Microsoft Fabric using Azure AI Studio. Seamlessly import your data from OneLake and choose from the most sophisticated large-scale language models hosted on Azure to build a custom AI environment for your data. All of this is possible only when you integrate and act on your data in Microsoft Fabric.