Businesswomen talking in meeting
·
5 min read

Dataverse at Microsoft Ignite Highlights

Happy Holidays! It was great meeting so many customers and partners at Microsoft Ignite. We shared lots of exciting announcements to help customers build and customize AI powered apps, agents, and automations for business needs across the organization. Many of you have started your AI journey with Microsoft Power Platform, which is built on a single, fully managed platform, Dataverse, that scales to meet the data needs of your enterprise. Here is a quick recap on how we continue to improve the platform experience: 

Confidently drive enterprise scale on the managed platform  

In the modern digital era, the opportunity for innovation by leveraging organizational data in AI-driven scenarios is at an unprecedented high. We recently introduced autonomous agents which can operate independently, dynamically plan, and learn over time. These new agents are rapidly increasing in complexity and widely distributed across the organization:  

  • Knowledge: connect to 1300+ Power Platform connectors from the Power Platform ecosystem.
  • Actions: evolving from retrieval only to automatically performing tasks on your behalf. 
  • Channels: published to multiple internal and external channels (beyond the boundaries of Microsoft 365 Copilot). 

As AI-driven scenarios continue to grow, IT admins are architecting the next wave of innovation in their enterprises by embracing managed platform features to meet modern data demands. With the security-first capabilities provided in Power Platform’s managed platform features, the possibilities for your business are endless. 

  • Managed Security: In the Power Platform Admin Center, Security Hub is the one stop shop for IT admins to enable AI-powered enterprise security management at scale. In Security Hub, IT admins can monitor their Security Score, which compiles many of our best practices and recommendations, evaluates how your tenant stacks against our recommendations into a single score, and suggests actions to boost your tenant’s security posture.
  • Managed Governance: In the Power Platform Admin Center, boost innovation while managing risk (Copilot Hub) and getting visibility into usage patterns for ROI analysis (Licensing Hub). Copilot Hub for agent administration is the central location for agent readiness, governance, and usage. Licensing Hub simplifies management of licensing, capacity, and overages. To learn more about Dataverse capacity management, read this blog on how we now enable PayGo for overages.  
  • Managed Operations: In the Power Platform Admin Center (for admins) and Maker Portal (for makers), we have Monitoring Hub to observe and maintain optimal operations by managing changes to production environments, detect and remediate incidents, and maintain business continuity.
Managed platform features to meet modern data demands. With the security-first capabilities provided in Power Platform managed platform including managed security, managed governance, and managed operations

Hear how we manage risk at Microsoft from MS IT. Read the full blog on how Power Platform Admin Center simplifies security and governance experience.  

Customize AI agents with enterprise knowledge in Copilot Studio 

As AI changes the ways we work, traditional workflows are getting reimagined as business process-oriented agents. In the same vein, enterprise data becomes valuable knowledge that agents can use to extend the range of functionality. However, even as makers are building more apps and agents than ever before, the majority of enterprise data still remains untapped. That’s changing with Copilot Studio, which ingests a range of data sources as “knowledge” to customize your agent including:  

  • Productivity data from Microsoft Graph (Outlook, SharePoint, Teams, etc.).
  • Line of Business data from Microsoft Dataverse and third-party data from external sources (Salesforce, Zendesk, ServiceNow). 
  • Pro-code assets from Microsoft Fabric, Azure AI Search, and Azure SQL. 
Makers can use knowledge from a variety of data sources to customize agents and extend the range of functionality in Microsoft Copilot Studio.

When incorporating these sources of enterprise data as knowledge, a customized agent is created, as demonstrated in Peppermint’s Personal Injury Claims agent demo. By integrating these data sources, businesses can accelerate business processes and drive transformative growth. 

Automate e-mail, doc, and image processing with GPT with AI Builder

Much of the new, incoming enterprise data is unstructured data from documents (see examples below), emails, images, audio, video, and text that will require analysis. And even today, most of our business processes remain heavily document centric. To address this challenge, AI Builder enables makers to optimize document workflows through process automation so you can gain insights from your unstructured data. 

Most of our business processes remain document centric including Telecom, Financial Services, Supply Chain, and more. Each industry has unique document processing needs.

To manage and process incoming documents, we announced new additions to enhance existing document process workflows (Custom Document with Azure Document Intelligence 4.0) for enhanced performance and confidence scoring.  

Additionally, we announced Document Processing with GPT. GPT can facilitate document processing which eliminates the need to label data and train new models. With AI Builder, makers can build and customize AI Prompts to:  

  • Extract information from documents. 
  • Detect the language used in text.  
  • Recognize text in an image or PDF (multimodality). 
GPT can facilitate document processing which eliminates the need to label data and train new models. Setting up a document processing workflow take hours/days instead of week/months!

And much more. To quickly get started, choose from a variety of prebuilt prompts in Prompt Library. Makers can also customize prompts with enterprise knowledge from Dataverse.  

Watch an end-to-end document processing workflow in:  

  • Power Automate (Customer support demo) to categorize, translate, redirect, draft, and respond to incoming customer service emails. 
  • Power Apps (Lost and Found demo) to describe a lost item based on an image and match that description to a database containing reports of lost items.  

AI Builder powered by GPT has transformed the way makers tap into their organization’s unstructured data by building end-to-end document processing workflows. AI Builder understands incoming documents, uses knowledge from Dataverse, and completes the workflow on your behalf. To learn more, read AI Builder’s announcements Ignite blog.  

Gain direct access to data for AI app development and analytics with Microsoft Power Platform and Snowflake   

We jointly announced an expanded partnership to enable bidirectional access between Microsoft Dataverse, the data layer of Power Platform, and Snowflake. This new Snowflake Power Platform connector enables developers and business users to:  

  • Leverage data in Snowflake directly from the Power Platform. Developers can now build end-to-end business applications with their data in Snowflake, removing the operational burden of managing custom workflows. Users can create a Dataverse virtual table from Snowflake (demo).  
  • Access data in Dynamics 365 and Power Platform from Snowflake to take full advantage of the AI Data Cloud to improve analytic insights and harness the power of enterprise AI.
With the new Snowflake bidirectional connector, makers can leverage Snowflake data in Dataverse. Conversely, makers can leverage D365 + Power Platform data in Snowflake.

See how it works and hear from SKF, a global manufacturing leader, on how they have implemented the connector and are experiencing faster time-to-value by removing some of the operational complexities of data integration. Together, Microsoft + Snowflake simplify data collaboration, enhance business insights, and harness the power of AI for our customers’ business needs. 

Recap 

Dataverse is the enterprise data platform for makers to add grounded enterprise data and make agents actionable. By building on Dataverse, makers can: 

  1. Enable scale and reduce risk through a managed platform including managed security, managed governance, and managed operations. 
  2. Incorporate knowledge from across your enterprise including productivity data, Line of Business data, and pro code assets.
  3. Easily build business process workflows with GPT from unstructured data including documents, emails, images, audio, video, and text.
  4. Tap into 1300+ Power Platform connectors (including the enhanced Snowflake connector) from the growing Power Platform ecosystem.  

Learn more about Microsoft Dataverse: 

Check out other great sessions to catch up on: 

Philip Li

Related posts