What's New > CLAIRE GPT > New features and enhancements
  

New features and enhancements

The February 2025 release of CLAIRE GPT includes the following new features and enhancements.

Discover and explore Data Integration assets

You can use data discovery and metadata exploration to discover and explore projects and folders in Data Integration as well as mappings and mapping tasks. For example, you can enter the following prompts for each journey:
You can enter similar prompts to explore folders within projects and mapping tasks.
For more information, see Discover assets from Data Integration and Explore metadata of assets from Data Integration.

Implied data discovery

CLAIRE GPT can identify and take action on the most relevant asset without performing data discovery.
For example, assuming that your data contains a single Accounts table, you can enter the following prompts:
For more information about data discovery, see Data discovery.

Next best prompts

If you have multiple assets with similar names and you want to identify and take action on the asset, the response lists all the relevant assets and their next best prompts. You can then use the next best prompt to confirm the asset in your response.
For more information about next best prompts, see Data discovery.

Data exploration

New source system
You can now preview sample data from Oracle, Microsoft Azure SQL Server, and Microsoft SQL Server source systems.
For more information about the source systems from which you can preview sample data, see Data exploration.
View asset details
You can view asset details of tables that do not have any relationships or references such as foreign key and primary key.
For example, you can enter the following prompts:
For more information about viewing asset details, see Data exploration.

Data quality

This release includes the following enhancements related to exploring the data quality of assets:
Data quality for system assets
Ask questions to explore the data quality of system assets in the catalog.
For example, you can enter the following prompts:
Lineage insights to diagnose the data quality of assets
When you ask questions to diagnose the data quality of data sets and data elements, the response now includes the upstream and downstream assets in the lineage of the asset. This helps you get insights into how upstream and downstream assets in the lineage of an asset affects the data quality scores of an asset.
For example, the following prompts return responses that include the lineage information of the assets in the lineage:
Enhanced recommendation of data quality rules
You can get data quality rule recommendations for data elements and data sets even if the asset is not associated with a data quality rule template.
The asset for which you want to generate rules may or may not be associated with a business term. If there are no business terms associated with the asset for which you want to generate rules, or if there are no data quality rule templates associated with the business terms, then CLAIRE GPT recommends rules directly from the existing data quality rule specification.

Explore the business lineage of system assets

You can ask questions to explore the business lineage of system business assets. While exploring the business lineage, you can ask questions to explore the target and parent assets in the lineage, assets contributing to the upstream and downstream lineage, assets that will be impacted by the system asset, and more.
For example, you can enter the following prompts:

Explore the analytical metrics of business assets

You can now ask questions to explore the analytics of business assets. You can also explore the analytical metrics of business assets based on their custom attributes.
For example, you can enter the following prompts to explore the analytics of business assets:
As another example, to explore the analytics of business assets based on their custom attributes, you can enter the following prompt where 'cost center' is a custom attribute with possible values as HR, Finance, and Marketing:

Clear conversation memory

After sending a few prompts and establishing context, you can now click Clear conversation memory to clear the memory of a conversation. This clears the context established in the conversation but retains the conversation history. After you clear the conversation memory, subsequent responses from CLAIRE GPT won't infer from the preceding prompts and responses in the conversation.
For more information about clearing conversation memory, see Clear conversation memory.

Response relevance based on user persona

CLAIRE GPT now sorts results in the response based on the user persona, such as a business user or a technical user.
For example, if you log in as a business user, the following prompts return responses that display business assets before technical assets:
This helps users discover relevant assets.
For more information about data discovery, see Data discovery.