Instant access to relevant and actionable insights is a game-changer. Recognizing this need, we are thrilled to introduce the latest addition to Cube’s expansive suite of tools: AI Assistant, now in preview.
This new feature empowers non-technical users to ask questions in natural language and receive trusted answers based on your existing investment into Cube’s universal semantic layer. Let's delve into the transformative capabilities of Cube's AI Assistant and how it can reshape your data accessibility and analytics processes.
The Evolution of Data Accessibility
Traditionally, accessing and interpreting complex data has been a domain reserved for data analysts and engineers, which has led to data teams being overwhelmed with requests for data. With Cube's AI Assistant, we aim to remove these barriers and enable self-serve data access for all users–regardless of their technical expertise.
AI Assistant leverages LLMs to facilitate natural language querying, enabling users to gain quick insights by simply asking questions in their own words. A business query such as "What were the sales figures for Q2?" can be answered without writing SQL or navigating a technical user interface.
Natural Language Processing for Seamless Interaction
Cube’s AI assistant does not generate SQL but instead generates an API request to the semantic layer - the text-to-semantic layer method not text-to-SQL. It does not rely on context from the semantic layer to enable it to write better SQL. It relies on the semantic layer to generate the correct SQL deterministically based upon the simple, constrained JSON format request. The AI Assistant generates this using retrieval augmented generation with metadata from the semantic layer and catalog. It requests known, understood and engineer-defined objects from the semantic layer. This results in greater consistency and transparency–without hallucinations–for both users and ease of debugging and validation for engineers.
Discover Modeled Data and Connect the Dots
The AI assistant can also answer metadata questions using the metadata stored in the Semantic Catalog. With this feature, you gain a holistic understanding of your data landscape. Users can easily discover available modeled data, downstream connected BI content, and upstream tables from the Semantic Catalog. This interconnected lineage provides deeper clarity and context, ensuring that all insights are framed within the broader operational data architecture of your enterprise.
Preview and Explore Interactive Charts
Visual representation is invaluable in data analysis, transforming raw numbers into understandable trends and actionable insights. Cube’s AI Assistant offers users the capability to preview charts rapidly, transforming queried data into visual formats such as bar and line charts. This feature not only makes data more digestible but also enhances the ability to identify patterns and outliers at a glance.
For users who need deeper exploration, the AI Assistant seamlessly integrates with Playground 2.0. With a click of a button, users can easily transition from the conversational interface with a simple chart preview to an interactive exploration environment. This deep dive enables further interrogation of data, the overlay of different variables, and the customization of visualizations to extract even richer insights tailored to specific analytical needs.
Democratizing Data Access within Organizations
Facilitating accurate, fast, and governed data is at the heart of Cube’s mission. By lowering the barrier for business users to work with enterprise data, the AI Assistant ensures that data teams aren’t overwhelmed by relatively simple ad-hoc data requests from stakeholders. At the same time, it leverages the data engineering work to build the semantic layer, which defines how the data model joins together and should be used. The meaning of data in your organization is still in the hands of your data engineers.
Stakeholders being able to safely, quickly and reliably self-serve with data—from marketing to finance to operations—leads to a more agile and responsive organization. Teams can align quickly on insights, pivot strategies based on real-time data, and enact changes with confidence, knowing they have the backing of reliable data.
Contact sales to learn more about AI Assistant and how it can lower the barrier to entry for everyone in your organization.