As enterprises deploy AI-powered analytics assistants across their data platforms, a familiar problem continues to emerge: the models can generate answers, but they often lack the business context needed to make those answers trustworthy. To address that challenge, DataHub has unveiled a major update to DataHub Cloud designed to provide analytics agents with a centralized layer of enterprise context.

The release positions DataHub Cloud as an intermediary between analytics assistants and enterprise data systems, supplying the metadata, business definitions, usage patterns, and institutional knowledge that AI systems frequently lack. The goal is to improve the accuracy and reliability of AI-generated analytics while reducing the risk of incorrect conclusions based on incomplete context.

Why Analytics Agents Struggle With Enterprise Data

The rapid adoption of AI assistants for data analysis has exposed a significant gap in enterprise AI architectures. While analytics agents can generate SQL queries and summarize results, they often do not understand which metrics are authoritative, which datasets are current, or how different teams define business concepts.

This challenge has become increasingly visible as organizations experiment with AI-powered analytics tools such as Databricks Genie and Snowflake Intelligence. Without access to trusted organizational context, these systems may select the wrong datasets, apply incorrect business logic, or rely on outdated information.

DataHub argues that solving this problem requires more than retrieval systems that simply expose schemas and metadata. Instead, AI agents need a continuously updated understanding of how data is actually used across the organization.

“Starting with Snowflake metadata alone, our analytics agent answered about half of our benchmark questions correctly,” said Ronald Angel. “After layering in DataHub Cloud as our context platform, including data product documentation, cross-source context and business meaning derived from our query history, we nearly doubled accuracy from close to 50% to around 90%.”

Building a Context Layer for Enterprise AI

The new release centers on the idea that enterprise context exists across many disconnected systems, from data catalogs and business intelligence tools to collaboration platforms and documentation repositories.

DataHub Cloud ingests information from more than 100 sources and combines structured metadata with unstructured organizational knowledge. The platform then organizes that information into what DataHub describes as a context graph that AI agents can query before generating responses.

A key component of the release is Context Intelligence, which transforms historical query activity into a searchable semantic knowledge base. Rather than relying solely on predefined semantic models, analytics agents can reference previously validated query patterns, joins, filters, and aggregation logic that have already been used successfully within the organization.

“Every system in the enterprise holds context, but DataHub Cloud is the platform built to unify it, and that makes it the natural foundation for AI agents,” said Björn Barrefors. “We have already seen it surface institutional knowledge that analysts would never have found on their own, while also flagging known data quality issues before the query ran. The next step is putting that context layer behind every data question our business asks.”

The platform also introduces Context Hub, a workspace where domain experts can review and refine AI-generated context, helping ensure that business definitions remain accurate as organizational data evolves.

Making AI Analytics More Trustworthy

The concept of a context layer has gained traction as enterprises move from AI experimentation toward production deployments. Analysts increasingly argue that data quality alone is insufficient; AI systems must also understand the meaning, lineage, and governance surrounding enterprise information.

“We turn years of query history into a living knowledge base, fuse in real-time operational signals and compound it with every expert correction from the field,” said Shirshanka Das. “Every change is timestamped and versioned; agents don’t just know the right answer, they know why it changed. That’s auditable context, and it’s how agents stop hallucinating and start earning trust.”

Beyond improving accuracy, DataHub says the platform can reduce AI inference costs by providing agents with pre-validated context rather than forcing them to process large volumes of raw metadata.

What Comes Next

As enterprises expand the use of AI-driven analytics, the industry is increasingly focused on a new challenge: ensuring that AI systems understand not just data, but the business meaning behind it. Context platforms are emerging as a critical layer between foundation models and enterprise information systems.

With this release, DataHub is betting that trusted context—not larger models—will become the key differentiator for enterprise analytics agents. If that assumption proves correct, context management may become as important to AI deployments as data management itself.