Key Highlights
- DataHub says AI analytics accuracy improved from nearly 50% to around 90% in benchmark testing.
- This update introduces Context Ingestion, Intelligence, Hub, and Activation features.
- The Platform aims to reduce incorrect outputs caused by missing or fragmented data context.
DataHub has introduced a major update to its cloud platform, launching a new AI context layer designed to improve the accuracy and reliability of analytics agents used by enterprises.
The new release, DataHub Cloud v1, acts as a bridge between analytics agents and enterprise data sources such as warehouses and data lakes.
Instead of relying only on raw metadata, the platform provides AI systems with business context, helping them better understand how data is defined, connected, and used inside organizations.
The launch comes as businesses increasingly adopt AI for analytics, while also struggling with inaccurate or misleading outputs. AI agents often generate incorrect responses because they lack context around metrics, business definitions, or updated data sources.
According to DataHub, its platform can significantly improve these outcomes. Ronald Angel, Product Manager of Data Platform at Miro, said an analytics agent initially answered only about half of the benchmark questions correctly using standard Snowflake metadata.
After integrating DataHub’s context platform, including documentation, cross-source information, and business meaning derived from query history, accuracy reportedly improved to nearly 90%.
New Features Focus on Trusted Context
The update includes several new capabilities aimed at making enterprise AI more dependable.
- Context Ingestion gathers fragmented information from structured systems and unstructured sources such as Notion and Confluence to create a unified context graph.
- Context Intelligence converts historical enterprise query data into a semantic knowledge base, allowing AI agents to retrieve validated query patterns, joins, and filters instead of relying solely on assaumptions.
- The platform also introduces Context Hub, where domain experts can review and refine AI-generated context before it is published, helping improve quality over time.
- Meanwhile, Context Activation enables organizations to integrate DataHub’s context into analytics agents and workflows through APIs, SDKs, and built-in tools.
Beyond accuracy, DataHub also claims the platform can lower AI operating costs by reducing the number of tokens needed for queries, as agents receive pre-validated context rather than processing raw data alone.
As organizations continue expanding AI adoption, the company believes trusted context will play a critical role in ensuring analytics agents produce more accurate and business-ready insights.
Read more news on Big Data.