SAP moves to acquire Dremio and Prior Labs, leveraging open source Iceberg and Polaris to make enterprise data AI-ready and accelerate tabular AI capabilities.
SAP SE has announced plans to acquire Dremio and Prior Labs GmbH, signalling strong enterprise validation of open-source data lakehouse standards as the foundation for AI-ready infrastructure. Financial terms were not disclosed.
The move centres on strengthening tabular AI capabilities—specifically enabling enterprise data stored in rows and columns to power agent-based AI workflows. A key driver behind the Dremio acquisition is its architecture built on Apache Iceberg and Apache Polaris, underscoring the growing role of open, interoperable systems in enterprise AI.
Iceberg enables schema evolution and faster querying of large datasets, while Polaris delivers robust metadata governance, including access control and change tracking—capabilities increasingly critical for AI deployment at scale.
SAP plans to integrate Dremio into its Business Data Cloud, enhancing Iceberg support, strengthening metadata management and improving overall AI data readiness.
“Enterprise AI doesn’t stall because the models aren’t good enough; it stalls because the data isn’t ready for AI agents,” said SAP Chief Technology Officer Philipp Herzig. “Dremio eliminates that bottleneck.”
Meanwhile, Prior Labs brings TabPFN-2.5, a specialised tabular AI model capable of processing up to 100,000 rows per task and detecting anomalies. Its distillation engine enables lightweight, faster, dataset-specific models. SAP will invest $1.17 billion in Prior Labs over four years, with the startup continuing to operate independently.
The acquisitions reflect a broader shift from proprietary data stacks to open-source-driven, AI-ready data ecosystems.
