SAP has completed its acquisition of Prior Labs, an artificial intelligence research company specializing in foundation models for structured enterprise data.
The financial terms of the acquisition were not disclosed. However, SAP plans to support Prior Labs with more than €1 billion in investment, providing the company with capital for computing infrastructure, hiring and long-term frontier AI research.
Prior Labs will continue operating under its existing brand, leadership team and research agenda. The company will also maintain its customer relationships, continue publishing research and make its models openly available with SAP’s support.
The transaction was completed approximately 18 months after Prior Labs was founded, giving the young company access to the resources and enterprise data environments required to pursue larger and longer-term research programs.
Prior Labs operates from Freiburg and Berlin, Germany, and New York. Its team includes researchers and engineers with experience at Google, Apple, Amazon, DeepMind, Meta, Microsoft Research, G-Research, Jane Street, Goldman Sachs and CERN.
The company was founded by Frank Hutter, Noah Hollmann and Sauraj Gambhir. Its advisers include AI researchers Bernhard Schölkopf and Turing Award winner Yann LeCun.
Prior Labs develops tabular foundation models, or TFMs, which are AI systems designed to work with the structured information commonly maintained by businesses.
Structured data is typically organized into rows, columns and defined fields. It includes information such as transactions, customer records, supplier activity, inventory levels, payment histories, operating metrics and financial results.
Although large language models have received substantial attention for their ability to work with text, businesses continue to rely heavily on structured datasets for forecasting, risk analysis and operational decision-making.
Traditional machine learning projects frequently require organizations to build and train a separate model for each dataset and prediction problem. This process can require specialized employees, significant computing resources and repeated rounds of preparation and testing.
Prior Labs’ TabPFN technology uses a pre-trained foundation model that can be applied directly to different structured datasets and prediction tasks.
The approach is intended to reduce the time and expertise required to build models for business use cases such as forecasting demand, predicting customer churn, identifying supplier risk and estimating payment delays.
Prior Labs said its TabPFN models have surpassed four million downloads and have been evaluated across hundreds of independent research projects.
The company recently introduced TabPFN-3-Thinking, its latest model for structured-data prediction tasks. Prior Labs describes the system as an enterprise-grade model capable of addressing a broad range of predictive applications.
Its technology is already being used by organizations including Hitachi and TD.
Hitachi is applying Prior Labs’ technology to help prevent train failures, while TD is using the models to support financial forecasting.
Researchers have also applied TabPFN to areas extending beyond conventional enterprise analytics, including pancreatic cancer diagnosis, wildfire prediction and the discovery of next-generation battery materials.
The acquisition will give Prior Labs access to SAP’s enterprise software ecosystem, customer base and global operating scale.
SAP’s systems manage substantial volumes of business information involving finance, supply chains, manufacturing, human resources, procurement and customer relationships.
Combining Prior Labs’ models with these enterprise data environments could enable the development of AI systems that identify patterns, generate predictions and support decisions within everyday business workflows.
For example, companies could use tabular foundation models to anticipate late customer payments, identify disruptions in a supply chain, forecast product demand or determine which equipment is most likely to require maintenance.
The technology may also support financial institutions evaluating risk, manufacturers monitoring production operations and healthcare organizations analyzing structured clinical information.
Prior Labs will remain independent in its research direction even as it gains access to SAP’s infrastructure and commercial reach.
This structure is intended to preserve the research culture that allowed the company to develop its technology while giving it the resources needed to conduct more ambitious work.
SAP’s backing will allow Prior Labs to pursue multi-year research initiatives that may have been difficult for an early-stage startup to finance independently.
These programs are expected to cover enterprise AI, scientific discovery, causal reasoning, relational data and agentic systems.
Causal AI seeks to identify relationships involving cause and effect rather than only recognizing statistical patterns. This capability could help businesses understand why an outcome occurred and how a particular decision might change future results.
Relational data research focuses on information connected across multiple entities and systems, such as the relationships among customers, suppliers, products, transactions and corporate departments.
Agentic systems are designed to carry out more complex sequences of tasks with limited human direction. Within enterprise environments, these systems could use structured data to analyze conditions, recommend actions and complete authorized workflows.
Prior Labs also plans to pursue larger research projects in fields such as medical data and materials science.
Scientific datasets are often structured but difficult to analyze because they can involve limited samples, large numbers of variables and complicated relationships among measurements.
Foundation models designed for this type of information could help researchers generate predictions more quickly and identify promising areas for further investigation.
The acquisition highlights SAP’s view that the largest untapped opportunity in enterprise artificial intelligence may be found in the structured information already running through business systems.
While large language models are effective at understanding and generating text, many of the decisions made by companies depend on numerical, categorical and transactional data.
Prior Labs’ technology is intended to make advanced machine learning more accessible for these datasets without requiring every organization to develop a separate model from the beginning.
SAP could eventually integrate the technology into software products used by its customers, although the companies did not provide a specific deployment schedule or announce individual product integrations.
Prior Labs will also continue working directly with customers and supporting open scientific collaboration.
The company’s ability to publish its research and make models openly available could help expand adoption among researchers, developers and enterprises outside SAP’s immediate customer base.
The transaction also represents a major investment in Europe’s artificial intelligence ecosystem.
By providing more than €1 billion in support while allowing Prior Labs to retain its brand and research independence, SAP aims to establish the company as one of Europe’s leading frontier AI laboratories.
The investment will help Prior Labs recruit additional researchers, expand its computing capacity and test its models across more complex enterprise and scientific environments.
For SAP, the acquisition strengthens its position in an area of AI closely connected with the company’s core enterprise software business.
Prior Labs gains the financial stability and data access required to advance its models, while SAP gains a specialized research organization focused on extracting value from the structured information used by businesses around the world.
KEY QUOTES:
“Eighteen months ago, Prior Labs was a research project. Today we’re beginning our next chapter as an AI lab with the resources to tackle problems we simply couldn’t before. Taking tabular foundation models to the next level requires better data environments, deployment surfaces and long-term research investment, and SAP is uniquely positioned to provide all of these.”
Frank Hutter, Co-Founder and CEO of Prior Labs
“Early on, SAP recognized that the greatest untapped opportunity in enterprise AI wasn’t large language models; it was AI built for the structured data that runs the world’s businesses. Prior Labs has defined the category of TFMs and has built the world’s strongest research team in this category, topping the public benchmarks since day one. Combining their frontier model work with enterprise data and customer reach is how we intend to lead this category globally.”
Philipp Herzig, Chief Technology Officer of SAP
