Google Cloud launches Cortex Framework v7 for SAP AI
Tue, 4th Aug 2026 (Today)
Google Cloud has launched Version 7 of its Cortex Framework, making the software generally available for customers working with SAP data.
The update targets companies that want to prepare ERP data for AI agents without changing core operational systems. It packages SAP data into reusable data products that can be deployed in BigQuery and registered in Knowledge Catalog.
Cortex Framework is designed to help businesses turn source records from SAP systems into data sets described in business terms rather than raw tables and fields. In Version 7, Google has added data product accelerators for SAP ERP and SAP Business Data Cloud, along with metadata support intended to make the information easier for AI systems to use.
The software can translate SAP tables into clearer business terminology, ingest custom fields and handle SAP-specific logic, including TCURX currency decimal shifts. That is intended to preserve data accuracy for analytics and large language model-based applications.
Modular deployment
A central change in the new release is the use of Dataform for pipeline deployment and orchestration. This gives customers version-controlled SQL, dependency management and a modular deployment model that retrieves only the tables required for a selected data product.
The structure is also intended to help companies running several SAP environments at once. The framework can compile and deploy data products for SAP ECC, SAP S/4HANA and SAP Business Data Cloud source systems in parallel, while automatically discovering custom SAP fields such as Z-fields.
The latest version also changes how data processing is handled in BigQuery. Cortex Framework now defaults to incremental loading and non-destructive schema updates, so only new or modified data is processed, which can reduce compute use and operating costs.
Customers that already use their own change data capture tools can connect those pipelines directly rather than relying on built-in change data capture processing. This gives users a bring-your-own approach for integration with current data estates.
SAP link
The release also expands interoperability with SAP Business Data Cloud Connect for BigQuery, which recently became generally available. Cortex Framework Version 7 now supports both standard and custom SAP Business Data Cloud data products.
That support is paired with solution samples that combine SAP Business Data Cloud and SAP ERP data. The examples are intended to help customers answer operational questions such as the state of sales pipelines and procurement performance.
Another addition is a set of agent skills for what Google described as an agentic data product builder. The feature is designed to let data and knowledge engineers use natural language prompts to generate custom data products, with the software handling parts of the build process and adapting to source system requirements and customisations.
Customers can also extend the framework while keeping custom models separate from the standard content delivered with Cortex. That separation is intended to let users apply software updates without affecting their own modifications.
Customer response
Google said it worked with its Professional Services Organisation and selected customers while developing the release. One of those customers was SLB.
"As an existing Cortex Framework customer, we are excited about the capabilities introduced with the latest v7 release. Bridging complex SAP ECC and S/4 environments with Google Cloud's AI capabilities requires a trusted data foundation with accurate business context. Based on the capabilities announced for Cortex Framework v7, including the agentic data product builder, we see significant potential to accelerate and scale the creation of enterprise data products for analytics and agentic AI use cases. The direction of the platform is particularly promising for organizations looking to make SAP data more accessible, semantically meaningful, and ready to support future reporting and AI initiatives," said Dr Jwan Ibrahim, Director, Data and Analytics, SLB.
The launch reflects a broader push by cloud providers to position data management tools as a foundation for AI software, particularly in large companies with established ERP estates. In this case, Google is targeting SAP users that need to expose operational data to AI systems while preserving business context and data governance.
Version 7 centres that effort on reusable data products, BigQuery-based processing and metadata registration in Knowledge Catalog, with support for SAP ECC, SAP S/4HANA and SAP Business Data Cloud in the same framework.