SAP bets $1.16 billion on German AI startup Prior Labs in tabular data play

Enterprise AI adoption remains limited despite the buzz around artificial intelligence. As OpenAI’s COO admitted in February, “we have not yet really seen AI penetrate enterprise business processes.” For SAP, whose stock has dropped significantly in 2026 amid the “SaaSpocalypse,” this presents both a challenge and an opportunity.

The German enterprise software giant is making its biggest bet yet on AI to protect its market position. The company has announced plans to acquire Prior Labs, an 18-month-old German AI startup, for an undisclosed amount and invest €1 billion ($1.16 billion) over four years to grow it into an AI lab focused on structured data.

What makes Prior Labs different

Prior Labs specializes in tabular foundation models (TFMs) – AI systems designed to work with data stored in tables and databases rather than text. This approach fits better with how enterprises actually store information, especially for SAP’s core products in accounting, HR, procurement, and expense management.

The startup has gained significant traction since its founding. Its TabPFN model series has been downloaded over three million times, and the founders – Frank Hutter, Noah Hollmann, and Sauraj Gambhir – are walking away with what sources describe as a “healthy exit” worth well over half a billion dollars in upfront cash.

SAP’s defensive AI strategy

While building its own AI capabilities, SAP is also restricting which AI agents can access its platform. The company has blocked OpenClaw and other unauthorized agent technologies, allowing only “SAP-endorsed architectures” through its API.

This includes:

• SAP’s own Joule Agents platform (still in beta)
• Nvidia’s NemoClaw agents, which work through SAP’s partnership with Nvidia’s Agent Toolkit
• Other explicitly authorized systems

This approach differs sharply from Salesforce, another enterprise giant hit by the SaaSpocalypse. Salesforce allows customers to choose their own agents, including OpenClaw, through its new Headless 360 architecture.

The broader context

SAP’s aggressive moves reflect the pressure traditional enterprise software companies face from AI disruption. As CFO Dominik Asam told CNBC in January, success depends on “how quickly we can as SAP actually also take on these technologies in our R&D portfolio to keep the relative economies of scale advantage.”

The company has already invested in several AI companies, including OpenAI rival Anthropic, as well as Aleph Alpha and Cohere (which plan to merge). SAP also developed its own SAP-RPT-1 model for working with relational data.

Prior Labs had previously raised $9.3 million in pre-seed funding led by Balderton Capital. Partner James Wise called the acquisition “one of Germany’s biggest ever venture outcomes.” The startup will operate as an independent unit while integrating with SAP’s existing AI infrastructure, including SAP AI Core and SAP Business Data Cloud.

For SAP, this acquisition represents a shortcut to building world-class AI capabilities for the structured data that powers business operations. Whether this defensive strategy can protect SAP’s market position as AI reshapes enterprise software remains to be seen.