SAP said it has signed a definitive agreement to acquire Prior Labs, a startup focused on tabular foundation models, and plans to invest more than €1 billion over the next four years to scale it into a frontier AI lab for structured business data.

What SAP is buying

According to SAP’s announcement, financial terms were not disclosed and the transaction is still subject to regulatory approval. Prior Labs is expected to keep operating as an independent entity, while SAP says its research will get a path into SAP AI Core, Business Data Cloud, and the Joule agent layer.

The core thesis is that standard large language models are not well suited to the tables, numbers, and statistical patterns inside enterprise systems. SAP says tabular foundation models are purpose-built for that kind of data and can support predictions around issues such as payment delays, supplier risk, upsell opportunities, and customer churn.

Why it matters

This is more notable than a routine AI partnership. SAP is making a large, explicit capital commitment around a specific model category that it believes fits enterprise software better than text-first LLMs alone.

Prior Labs said its open-source TabPFN work will remain open after the deal, and SAP separately said the lab will continue operating independently to preserve research speed. The conservative takeaway is that major software vendors are starting to treat structured-data AI as its own battleground, with native reasoning over business tables becoming a strategic layer rather than just another chatbot feature.