Ataccama brings data trust to Apache Ossie, enabling AI agents to verify data quality before acting

Open-source converter helps AI agents understand what enterprise data means and whether it is fit for use, with trust signals that span source systems and data platforms

BOSTON, Sept. 09, 2026 (GLOBE NEWSWIRE) -- Ataccama, the data trust company, will open-source a converter that carries governed business definitions and current data quality signals into Apache Ossie (Incubating). Apache Ossie is the open semantic specification originally introduced by Snowflake as Open Semantic Interchange (OSI) and now an Apache Software Foundation incubating project. The converter lets AI agents and analytics tools read what enterprise data means and whether it is currently reliable enough to use. Because Ataccama evaluates quality in the source systems where data originates as well as across cloud data platforms, those trust signals travel across AI and analytics environments without requiring the underlying data to live on a single platform.

The distinction between data’s meaning and reliability is becoming critical as AI takes on more consequential work. Semantic layers give agents a consistent understanding of business data so they know, for example, what revenue means, which records represent a customer, and how to calculate a metric. However, a perfectly defined metric can still produce the wrong answer if the underlying data is incomplete, late, or failing a critical quality check. As AI moves from helping people interpret information to recommending decisions and taking action with less human review, the quality of the underlying data has to become part of the context AI uses to reason. 

“An open semantic standard promises that enterprises can define their business once and carry that understanding across the tools that use it,” said Josh Klahr, Head of Product Management at Snowflake. “Ataccama is extending that portability to data quality, giving organizations a consistent way to make those signals available across platforms rather than rebuilding them for each environment. That is an important contribution as enterprises connect more of their data to AI.”

Ataccama combines the two capabilities this architecture requires: governed business context that establishes what data means, and continuously evaluated data quality that shows whether it is fit to use. The converter brings both from Ataccama ONE into Apache Ossie, and Ataccama's MCP Server gives agents access to deeper evidence when they need to reason about a quality issue.

Those signals show whether data is currently meeting its quality threshold and whether Ataccama has identified active quality issues, eliminating the need for quality indicators to live in separate dashboards waiting for someone to spot a problem. An agent can use those signals at the point of use, factoring data quality into its reasoning before producing an answer or taking action. The signals travel with the semantic definition, making the same view of data quality available wherever that definition is used.

“AI exacerbates the impact of incorrect data,” said Jessica Smith, Chief Product Officer at Ataccama. “In the BI era, bad data typically got caught because it conflicted with something already known – another report, a trusted baseline, someone's memory of the last board deck. AI agents need that same level of operational perspective. An agent may know what revenue means, but not whether the revenue data its reading is complete or current, and it will still produce a convincing answer. Opening our converter to Apache Ossie will help agents inherit the context that people used to supply, and the critical ability to know whether data should be trusted, thanks to Ataccama's data quality signals now available in the semantic layer itself.”

Ataccama will open-source the converter so quality context can move with a semantic definition wherever it goes, rather than becoming another proprietary layer that organizations have to rebuild for every platform. The converter will join contributions to Apache Ossie from across the data ecosystem, including Snowflake, Databricks, and dbt.

Ataccama’s new converter will enable data trust for AI through:

• AI-readable quality warnings. Warn agents when data falls below its accepted quality threshold, so they can qualify or investigate questionable data before it informs an answer or action.
• Continuously updated trust signals. Fresh quality signals from Ataccama's checks, so agents work from the data's current state rather than a point-in-time certification.
• Trust evidence on demand through MCP. Agents receive an immediate quality signal through the semantic model. They can investigate further through Ataccama's MCP Server, retrieving details such as which quality checks failed and the records affected when they need to understand or act on an issue.
• Portable trust from source to AI. Evaluate quality in the source systems where data originates and across cloud data platforms, then carry those signals into AI through Ossie-compliant YAML. Enterprises can preserve a consistent view of quality across heterogeneous data estates without first moving everything into one cloud or rebuilding quality context around every platform.

The converter is the latest step in Ataccama’s work to make trusted enterprise data accessible and actionable for AI, extending how agents can access, understand, and assess enterprise data through capabilities including its MCP Server and ONE AI Agent.

Join Ataccama’s Ariel Pohoryles and Matěj Matoulek alongside Snowflake’s Josh Klahr on September 15 for “The 3-layer formula for trusted AI agents” to see how semantics, context, and live data quality signals work together to support trusted AI. Register for the webinar.

Read “Bringing the trust layer to your semantic layer: Ataccama's open-source converter for Apache Ossie (Incubating)” to learn more. 

About Ataccama
Ataccama provides the only end-to-end agentic data trust platform that helps organizations accelerate AI, reduce risk, and modernize data at enterprise scale. Ataccama ONE sits between enterprise data and AI orchestration, ensuring trusted data powers every model, agent, and business decision. The platform unifies data quality, observability, catalog, lineage, reference data management, and master data management in a single solution, while the embedded ONE AI Agent acts as a digital data steward that automates repetitive data management work. Recognized as a Leader in the 2026 Gartner® Magic Quadrant™ for Augmented Data Quality, the 2026 Forrester Wave™ for Data Quality Solutions, and the 2025 Gartner® Magic Quadrant™ for Data and Analytics Governance, Ataccama helps the world's leading enterprises trust their data so they can accelerate AI. Learn more at www.ataccama.com.

Media contact

Lauren Ruth
Director of Global Communications
lauren.ruth@ataccama.com