SAP completes its billion euro bet on a tiny German AI lab nobody outside academia had heard of 18 months ago

[WildManSteve40]

SAP officially completed its acquisition of Prior Labs, a Freiburg based startup founded just 18 months ago, and is committing more than 1 billion euros over the next four years to scale it into what SAP is calling a globally leading frontier AI lab for structured business data. The deal was first announced back in May and closed on July 17 following regulatory approval, with Prior Labs continuing to operate as an independent entity under its own brand, leadership and research agenda

What makes this deal unusual is the focus, Prior Labs is not building large language models or chatbots at all, it pioneered something called Tabular Foundation Models, AI purpose built specifically for structured data like spreadsheets, ERP tables and operational databases rather than free form text. Its TabPFN model line was published in Nature and reportedly set state of the art results across hundreds of independent academic benchmark studies before the company had even raised significant funding, having previously taken in just 9 million euros in a pre seed round

Founders Frank Hutter, Noah Hollmann and Sauraj Gambhir framed the deal as giving them resources previously out of reach for a company this young, enabling ambitious research into causality, relational data and agentic systems built around structured business information rather than text. It is one of the fastest European deep tech exits in recent memory, and a notable bet from SAP that the biggest untapped opportunity in enterprise AI is not another chatbot but better predictions drawn directly from the operational data that already runs most businesses

SAP Completes Acquisition of Prior Labsnews.sap.com

[EntangledOne]

Going from a research paper to a billion euro acquisition in 18 months is an absurdly fast timeline even by current AI industry standards

[error.404]

Tabular foundation models are such an underrated category, most of the AI conversation is dominated by chatbots when this structured data problem is arguably more valuable for actual businesses

[Rogue Di]

9 million euro pre seed to over a billion in committed investment is one of the biggest funding jumps I have seen for a company this young

[Tyler_16]

SAP keeping Prior Labs independent rather than fully absorbing it seems like the right call, that kind of research culture rarely survives a full corporate integration

[Foundry20]

This is a really smart complementary bet next to their existing SAP-RPT-1 work, makes sense to buy proven talent rather than build this internally from scratch

[AuroraHermit]

Nice to see a notable European AI exit for once, the continent's deep tech scene usually gets compared unfavorably to the US and China on this kind of outcome

[Woven Sarah]

The causality and relational data research angle is the part that actually excites me here, that is hard unsolved territory

[WearyCoder]

Curious how this actually changes day to day SAP products for customers rather than just being a research investment on paper

[TheGame92]

Publishing in Nature before even raising serious funding says a lot about how strong the underlying science was from day one

[Maya98]

Would like to see a real world case study of TabPFN actually outperforming a traditional AutoML pipeline on a real enterprise dataset