Pega Platform™ '24.1 expands the scope of services based on Microsoft Azure OpenAI Service. As a result, you can use a number of new out-of-the-box Pega GenAI™ functionalities to improve development and meet your business needs.

For more information, see Integrating with Pega GenAI.


Create a new application from a Pega GenAI-powered Blueprint

You can now create the basic structure of your application in a rapid, low-code way by using Pega GenAI™ Blueprint. After you describe your business needs and the purpose of your application, Blueprint generates a comprehensive set of Case Types, data objects, and Personas tailored to your specific needs. You can then import the Blueprint directly into the New Application wizard and continue development in Pega Platform. As a result, you save development time by significantly reducing the planning phase.

The following video shows the process of designing an application with Blueprint and then importing the design with the New Application wizard:

For more information, see and Creating a new application from a Blueprint.

Pega GenAI™ Autopilot for enhanced application development

You can now use Pega GenAI Autopilot in App Studio to help you achieve faster, more intuitive, and more assisted application development. Select one of the ready-made conversation starters or enter your own questions to receive detailed information and steps on how to create or edit elements of your application. Apart from conversing with Autopilot, you can also store your conversations for future reference, rename the conversations for easy identification, and search through them. For further enhancement of the development process, Autopilot provides links to Pega knowledge resources and action buttons that launch the authoring experience, so that you can obtain information and create new objects in App Studio even faster.

The following video shows a sample conversation with Autopilot:

For more information, see Pega GenAI Autopilot in application development.

To accelerate and automate your Case Type creation experience, you can now use Autopilot in the following ways:
Creation of Embedded Data and data reference fields
When you create a Case Type by using Autopilot, you can choose from AI-generated Embedded Data and data reference fields. To adjust the fields to your unique business requirements, you can edit fields' names and configurations, as well as edit individual fields within a data object.

For more information, see Creating a Case Type with Autopilot.

Visual preview of a Case Type workflow
During a Case Type creation, Autopilot suggests Stages and Steps in your workflow, that you can preview and edit before submission, as shown in the following figure:

Sample Stages and Steps in a Health Insurance Case Type suggested by Autopilot.

Generating a Case workflow with Autopilot

For more information, see Creating a Case Type with Autopilot.

Automated creation of Views
When you populate an empty View with fields in a Step configuration pane, you can use Autopilot to suggest the fields for your form based on the names of the View, Case Type, and Stage, as shown in the following figure:

A comparison of forms before and after adding fields suggested by Autopilot.

Populating a form with Autopilot

For more information, see Configuring forms.

Generation of email correspondence
To automate generation of your email correspondence, you can now use Autopilot in a Send Email Step in your Case Type. Based on the names of the Case Type, Stage, and Step, Autopilot suggests message content, that you can modify and regenerate if required, as the following figure shows:

Send Email Step with a message generated by Autopilot.

Generating email content with Autopilot

For more information, see step 6 in Sending automatic emails from Cases.

For more information about Autopilot in Case Management, see Autopilot features in Case Type creation.

Pega GenAI™ Coach available in Constellation

In Constellation applications, you can now add and configure Pega GenAI™ Coach, a generative AI-powered mentor for Pega solutions that proactively advises users to help them achieve optimal outcomes. Coaches can be easily configured to ensure that each Coach is perfectly tailored to an organization’s objectives and their employees’ specific needs. You edit the definitions, instructions, and data sources, and Pega GenAI Coach runs by configuration at run time.

In the following example, Pega GenAI Coach is configured to summarize Case data at run time on a tab. Pega GenAI Coach condenses Case data and information into a single paragraph and makes it easier for users to review progress and recent changes to the Case:

Default Pega GenAI Coach provides a Case Summary as a response to the initial instruction.

Pega GenAI Coach providing a Case summary

For more information, see Enabling Pega GenAI Coach.

Generative AI integration of your client-managed deployment with Pega GenAI™ PremBridge

You can now enable generative AI capabilities in your Pega Platform applications on client-managed deployments by using Pega GenAI PremBridge. You request a Pega Cloud instance with the Pega GenAI PremBridge application installed, and then you can securely unlock the full potential of Pega GenAI.

For more information, see Integrating client-managed deployments with Pega GenAI.

Using Pega GenAI™ in data management and integration

Creating a Data object with AI

Instead of manually entering the name and fields for the Data object that you create, you can now select AI-suggested options. Pega GenAI™ suggests Data object names that are based on the application name, as well as Data object fields that are based on the application name and the Data object name.

For more information, see Creating a data object with AI.

Adding sample records to Data Objects with AI

You can now use Pega GenAI to populate the fields in a Data object with sample records.

For more information, see the Adding sample records locally to Data Objects with AI section in Configuring a basic data object.

Mapping external API responses for Data objects with AI

You can now use Pega GenAI to help you to map fields in the visual data mapper. Pega GenAI first normalizes the names of unmapped fields by ignoring case, special characters, and generic prefixes of fields. Pega GenAI then suggests a mapping for the field names that obviously match each other.

For more information, see Mapping fields for Data pages with AI.

Generative AI integration in Case Management

Use Generative AI to create Approve/Reject and Send Email Steps during Case Type creation

You can now create Approve/Reject and Send Email Steps using generative AI during Case Type creation.

Pega GenAI can determine a requirement for Approve/Reject and Send Email Steps in a new Case Type from a prompt and suggest them in the Case life cycle. This new feature helps users to quickly and efficiently build their applications using a variety of Pega GenAI suggested Steps.

Steps generated by AI now include Send email and Approve/Reject. These can be modified before finalizing Case Type creation.

AI-suggested Send email and Approve/Reject Steps

Pega GenAI Agent Trainer

You can now use Pega GenAI™ Agent Trainer to emulate customer interactions within your Digital Messaging Channel, enhancing CSR training and feedback. By establishing a connection with your Pega Platform application, you can simulate chat sessions with sample customers, define scenarios, and set up personas for comprehensive training. This feature improves CSR performance and customer satisfaction by providing real-time insights and access to saved data.

For more information, see Setting up a Pega GenAI Agent Trainer for Pega Customer Service, Adding a Pega GenAI Agent Trainer connection, Configuring a Pega GenAI Agent Trainer simulation, and Running a Pega GenAI Agent Trainer simulation.

Generative AI integration in Pega Intelligent Virtual Assistant (IVA)

AI-generated training data for response commands

Use Pega GenAI to automatically generate representative user utterances to help trigger a response command. You can then use this training data to instantly rebuild your system's text analytics model to improve the chatbot responsiveness during a chat interaction.

Sample utterances generated using AI related to asking for the help topic.

Sample utterances generated with Pega GenAI for a response command

For more information, see Generating training data for response commands with AI.

Provide additional context to AI suggestions

You can now refine AI-generated suggestions by adding extra contextual information that best fits your business use case. For example, when Pega GenAI suggests Stages and Steps for your Case Type, you can then add text that explains your business use case better, and then regenerate the suggestions.

For more information about generating assets with Pega GenAI, see Creating a Case Type with Autopilot, Creating Personas with AI, Creating a data object with AI, Creating a Case Type with Autopilot.

Dialog box for providing more context for the list of AI-generated Case Stages.

Regenerating AI suggestions with more context

Additional customization options for Connect Generative AI Rules

Customize the system prompt
You can now customize the message that the system includes with every prompt to Pega GenAI. By changing the system prompt, you can refine the generative AI suggestions for a particular business use case, for example, to list keywords to exclude.

The system prompt gives the generative AI model instructions, such as expected behavior and additional context to reference when generating a response. With the system prompt, you can define the personality of the AI assistant, and include or exclude specific answers.

Set the language of the Pega GenAI suggestions to the current operator locale
You can configure Pega GenAI to always provide responses in the language based on the operator's default locale. Otherwise, the language of each response is based on the language of the prompt.

For more information, see Creating a Connect Generative AI Rule.

AI-supported localization

App Studio in Constellation applications now includes tools for pre-translating your application with generative AI. The new Localize with GenAI option on the Localization landing page helps you jump-start the localization process and provides important insights about the look and feel of the translated application early in the design process.

For more information, see Pre-translating your application with GenAI in Constellation.

Process for creating a pre-translated the package.

Creating a pre-translated language pack

This feature is a Constellation counterpart to the Pega GenAI localization toolset made available with Pega Platform version '23. As part of that update, App Studio gained a new Localization landing page that includes the option to export machine translated text or import external translation files.

For more information about using machine translation to speed up the localization process, see Localizing your traditional application with AI.

Gen AI provides a sample translation from French to Czech.

The Localization landing page with AI-supported machine translation