SAP invests €1B+ in Prior Labs AI acquisition in Freiburg
Updated on June 05, 2026 4 minutes read
SAP has agreed to acquire the Freiburg-based AI startup Prior Labs and commit more than €1 billion over four years to expand its enterprise AI capabilities. The announcement was made on 4 May 2026 and marks one of the largest European AI investments in recent years. The acquisition is still subject to regulatory approval and is expected to close in 2026.
Rather than focusing on general-purpose chat systems, SAP is positioning this investment around structured data intelligence for enterprise systems.
What happened
On 4 May 2026, SAP SE announced a definitive agreement to acquire Prior Labs, an AI startup based in Freiburg, Germany. The company is known for its work on Tabular Foundation Models, which are designed to learn directly from structured business data such as spreadsheets and databases.
Alongside the acquisition, SAP committed more than €1 billion in planned investment over a four-year period to scale Prior Labs into a frontier AI research unit within its enterprise ecosystem. The goal is to strengthen SAP’s position in enterprise AI by focusing on data formats that dominate business environments.
Prior Labs was founded in 2024 and gained recognition for its TabPFN model family, which applies transformer-based approaches to tabular datasets. Unlike large language models that are optimized for text, these systems are designed to improve predictions on structured datasets commonly used in finance, logistics, and enterprise planning.
The deal is expected to close in Q2 or Q3 2026, pending regulatory approval and standard closing conditions.
Why it matters
This acquisition highlights a shift in enterprise AI priorities. While much of the industry focus has been on chatbots and generative text models, SAP is investing in AI systems optimized for structured enterprise data.
Most real-world business data is stored in tables rather than text. Financial records, supply chain systems, HR databases, and ERP platforms all rely heavily on structured formats. Traditional large language models are not always reliable in these environments.
By investing in tabular foundation models, SAP is targeting improved forecasting, classification, and decision support directly inside enterprise systems. This could reduce reliance on manual data pipelines and improve automation in business operations.
For learners and developers, this shift suggests growing demand for skills in applied machine learning, data engineering, and enterprise system integration rather than only prompt-based AI usage.
Key numbers
- €1+ billion planned investment by SAP over four years
- 4-year strategic development horizon for Prior Labs integration
- Founded in 2024 by the Prior Labs team
- Expected deal closure in Q2 or Q3 2026
- Millions of downloads reported for TabPFN research models
- Acquisition announced on 4 May 2026
Context
SAP has been steadily expanding its artificial intelligence strategy across its enterprise software portfolio. This includes tools for predictive analytics, automation, and AI-assisted business workflows.
Prior Labs represents a more specialized direction within AI research. Instead of focusing on language generation, it focuses on structured data learning, which is a historically difficult problem in machine learning.
In the broader industry, cloud providers and enterprise software companies are competing to embed AI directly into business infrastructure. This includes CRM systems, ERP platforms, and supply chain management tools. SAP’s move aligns with a wider trend where companies are investing in domain-specific AI systems rather than general-purpose models alone.
What’s next
Once regulatory approval is complete, Prior Labs is expected to operate as an independent research unit under SAP. The focus will likely be on scaling tabular foundation models for production use in enterprise environments.
SAP is expected to integrate these models into its Business Data Cloud, SAP AI Core, and its AI assistant ecosystem. This could enable more advanced forecasting, anomaly detection, and automated decision-making across enterprise workflows.
For developers, this may lead to new APIs and tooling that connect structured enterprise data directly to machine learning models without traditional preprocessing pipelines.
How to go deeper
If you want to understand the skills behind this shift in enterprise AI, you can explore:
- Data Science Bootcamp to learn machine learning on structured data
- Web Development Bootcamp to understand how AI integrates into modern applications
- Cybersecurity Bootcamp to learn how enterprise systems protect sensitive data