Share via


Plan and prepare for Microsoft Dataverse in 2024 release wave 2

Important

The 2024 release wave 2 plan covers all new functionalities planned to be delivered to market from October 2024 to March 2025. In this article, you'll find the product overview and what's new and planned for Microsoft Dataverse.

Overview

Microsoft Dataverse is a low-code data platform that you can use to build scalable and interconnected applications, automations, and agents with copilot capabilities. Here are some key capabilities of Dataverse:

  • Do more with your data. Easily use data and metadata from different sources for generative AI workflows and copilot applications.

  • Enrich your copilot experience. Integrate, search, and act on the data you use every day.

  • Enrich your copilot experience. Create dynamic prompts with grounded enterprise data by using Azure OpenAI large language models (LLM) and connect to external sources.

  • Build custom actions. Build custom connectors, plugins, and custom APIs that package business logic and services into your business process.

  • Proactive data security. Assess, act, and safeguard your data with advanced security features.

  • Fully managed, fully integrated. Get built-in support for enterprise application lifecycle management across the Power Platform ecosystem.

  • End-to-end security. Protect your data with a robust security management infrastructure that provides critical security and compliance capabilities. Get advanced encryption, rich access control, and deep integration with Microsoft Entra ID. Build on trust with advanced security features and interoperation with Microsoft Sentinel.

  • Compliance capabilities. Get audit logs of all AI data access with Microsoft Purview.

RSS Subscription Updates to Microsoft Dataverse 2024 release wave 2

Investment areas

Investment areas

Build your low code hyperscale backend on Dataverse
This semester, you see faster low-code experiences for apps built on Dataverse. You can create reusable low-code business logic in Dataverse that you catalog so other makers in your organization can discover and use your apps. To increase the value of maker investments in your organization, use these reusable actions by copying and pasting Power Fx logic from a canvas app into Dataverse actions. You can also call connector actions directly to run external business logic. You see these actions available across Power Platform.

Data workspace
As you start using Copilot, your enterprise data plays a key role in adopting reliable Copilot-enabled applications.

With data workspace features, Dynamics 365 and Power Platform customers can use AI-assisted mapping to simplify integrating external data into existing Dataverse tables. This feature streamlines data management. It offers assisted mapping for standard dataflows and lets you bypass business logic to speed up data migration. You can load data from Excel directly into a Dataverse table. You can also use Fabric One Lake as a data source without ETL or copying data. These options simplify your data import and export strategy.

The product also provides integrated experiences to upgrade from Export to Data Lake, maintain security role privileges for better ALM solution imports, and mask sensitive data with column-level security. Security features prevent data exfiltration with app access and user-level controls. You can scan files and attachments for viruses and malware. Modernizing sharing apps and data access, along with sharing system views with security roles, focuses on collaboration and accessibility.

Extend your copilots with knowledge and actions
Dataverse lets you extend Copilot for Microsoft 365 and custom agents you create in Microsoft Copilot Studio. Use Copilot plugins to add tools (formerly known as actions) and connect to enterprise knowledge sources. These plugins help improve the reliability and quality of responses.

Adding tools and specific knowledge is the best way to extend your copilot. You can connect to external systems, write specific instructions, or set up specific references. Plugins are the most popular version of an action. Use existing plugins in Copilot Studio or create your own. Set specific descriptions for your plugins to help Copilot call them. To change how you or your users interact with Copilot, create prompts. Prompts give specific instructions and can ground Dataverse data into Copilot.

In Microsoft Copilot Studio, the knowledge center uses Dataverse as the central place where you add and manage all your data sources and knowledge sources across all three clouds with your copilot:

  • Operational data (Dataverse)
  • Productivity data (Microsoft 365)
  • Analytical insights (Microsoft Fabric)

Copilot Studio includes built-in Dataverse. Dataverse connectors and plug-ins work as APIs or as an MCP server. These options offer two distinct plugin systems. Each system has different architecture and purpose. Copilot plugins use APIs to extend the conversational abilities of AI copilots and agents. In contrast, Dataverse plug-ins are C# components that run in a sandboxed process within Dataverse. Use these plug-ins to enforce business logic directly within Dataverse.

With simple configurations and natural language, you can use the knowledge center to connect all your Copilot experiences to the knowledge sources you choose. This connection helps customers get more value from their enterprise data. Besides Microsoft data sources, you can also provide other knowledge sources, such as Salesforce, ServiceNow, and Zendesk, to the custom Copilot through this knowledge experience.

Improved enterprise experiences in Power Platform
When you improve the experiences for building apps, flows, and solutions by using templates and standard components, and by enhancing how you connect, map, and use data, enterprise organizations find more engaging experiences across Power Platform and Microsoft Dataverse.

To learn more about the entire set of capabilities being delivered during this release wave, check out the release plan for Microsoft Dataverse below:

For application administrators

User-impacting features to the user experience enabled automatically
User-impacting features should be reviewed by application administrators. This facilitates release change management and enables successful onboarding of new capabilities released to market. For the complete list, look for all features tagged "Users, automatically" in the release plan.

Features that must be enabled by application administrators
This release wave contains features that must be enabled or configured by administrators, makers, or business analysts to be available for their users. For the complete list, look for all features tagged "Users by admins, makers, or analysts" in the release plan.