SAP closes Prior Labs acquisition with €1B-plus backing

SAP has completed its acquisition of Prior Labs, giving the 18-month-old AI startup more than €1 billion to expand its frontier research on enterprise data. The deal keeps Prior Labs operating under its own brand while linking its tabular foundation models to SAP’s enterprise data reach.

Why it matters: - SAP’s backing gives Prior Labs the capital and distribution needed to push tabular foundation models into more enterprise use cases. - The deal adds long-term support for frontier AI research focused on structured business data, not just large language models. - Prior Labs plans to keep publishing research and make its models openly available, which keeps the work visible to the wider AI community.

What happened: - SAP completed its acquisition of Prior Labs on July 17, 2026. - The transaction values the company’s next phase at more than €1 billion in SAP support. - Prior Labs was founded 18 months ago and will continue under its own brand, leadership, research agenda and customer relationships. - SAP’s support will cover infrastructure, hiring and long-term frontier research.

The details: - Prior Labs built tabular foundation models, or TFMs, for enterprise data. - TabPFN uses one pre-trained foundation model to handle prediction tasks from structured data. - The model is used for jobs such as payment delays, churn, supplier risk and demand forecasting. - Prior Labs says its technology has helped prevent train failures with Hitachi and improve financial forecasting with TD. - The company says its models have also been used across hundreds of published research projects, including pancreatic cancer diagnosis, wildfire prediction and next-generation battery materials. - TabPFN-3-Thinking is the company’s latest model and is described as state-of-the-art and enterprise-grade for prediction tasks. - Prior Labs operates from Freiburg and Berlin, Germany, and New York. - The company was founded by Frank Hutter, Noah Hollmann and Sauraj Gambhir. - Prior Labs is advised by Bernhard Schölkopf and Yann LeCun. - Its TabPFN series was published in Nature and has 4 million downloads. - The company says its models set the state of the art on tabular benchmarks across hundreds of independent studies.

Between the lines: - SAP is signaling that enterprise AI value may come from structured data workflows, where many businesses already store their most useful information. - The acquisition also gives Prior Labs a stronger path to scale research that would be hard to fund as a young standalone lab. - Frank Hutter said the deal gives Prior Labs the resources to tackle problems it could not before. - Hutter said better data environments, deployment surfaces and long-term research investment are needed to take tabular foundation models to the next level. - SAP CTO Philipp Herzig said the company believes the biggest untapped opportunity in enterprise AI is AI built for structured data.

What’s next: - Prior Labs will use SAP’s enterprise data ecosystem and global scale to build the next generation of foundation models for enterprise data. - The company plans multi-year research programs in enterprise AI, scientific discovery, causality, relational data and agentic systems. - Prior Labs also wants to pursue longer-shot work in medical data and materials science. - SAP and Prior Labs will try to turn research leadership in TFMs into broader commercial deployment across industries.

The bottom line: - SAP is betting that the next frontier in enterprise AI is not only generative text, but foundation models trained for the structured data that runs businesses.