The Gio™ AI Assistant supports conversational copilot, data cleansing and classification, spatial enrichment, and natural language rule creation in Data Integrity Suite. These functions automate manual processes including pipeline creation and rule recommendation. Each agent follows principles of trust and confidentiality, with transparent explanations for its recommendations. Gio™ AI Assistant is built on the AI Agent as a Service framework, which lets developers register, configure, and deploy AI agents across services.
The main functionalities supported by Gio™ AI Assistant are:
- Conversational Copilot: Supports natural language
interactions for cross-module operations. Questions are
converted into search actions to find relevant datasets
and assets. Conversations are saved for easy revisit.
- Select a starter card, such as "Find Something", or type your question into the "Ask Gio™ AI Assistant" input box.
- To view previous interactions, click the History button.
- To start a new conversation, click New Chat.
- Example use cases:
- Dataset discovery: Search across assets using natural language. For example, "Find the latest 'Sales Bookings' data from previous quarters".
- Asset exploration: Explore assets related to specific topics or policies, such as GDPR, by asking for a list of assets tagged accordingly. For example, "Which assets are tagged to a GDPR policy?".
- Lineage visualization: Understand data dependencies by asking about upstream and downstream lineage. Gio™ AI Assistant returns a lineage summary showing data sources feeding into your dataset, along with key details on the datasets and fields involved. You can also view the complete lineage diagram for deeper analysis.
- Data quality and trustworthiness: Get embedded data quality insights, including overall quality scores, data profiling information such as volume and refresh dates, and trustworthiness assessments. This information is sourced from your data catalog, enabling you to verify details and make informed decisions. For example, "Can I trust the 'telco_subscriptions' dataset?".
- Data cleansing agent: Standardizes and cleanses data to improve quality and consistency across datasets.
- Data classification agent: Automatically classifies and categorizes data based on content and context.
- Spatial enrichment agent: Enhances data with geographical and location-based information.
- Data quality discovery and creation agent: Discovers data quality issues and helps create rules to maintain data integrity.
-
Pipeline replication: Enables real-time data synchronization between systems to ensure consistency and support timely, informed decisions.
Each agent automates manual tasks, streamlines operations, and reduces user workload.