SAP Acquires German AI Lab Prior Labs for $1.16B to Build Europe's Frontier AI Research Center

SAP announced on May 5 that it will acquire Prior Labs, the Freiburg, Germany-based startup that pioneered Tabular Foundation Models, with a commitment of more than €1 billion (approximately $1.16 billion) invested over four years. The deal — the largest European AI research acquisition announced in 2026 — gives SAP ownership of Prior Labs' TabPFN model series, which holds state-of-the-art performance on structured tabular data benchmarks across hundreds of independent academic studies. The transaction is expected to close in Q2 or Q3 of 2026, subject to regulatory approval. Prior Labs will continue operating as an independent legal entity, retaining its brand, research team, and headquarters in Freiburg, with additional offices in Berlin and New York.

What Prior Labs Built — and Why SAP Needed It

Prior Labs was founded in late 2024 by Frank Hutter, Noah Hollmann, and Sauraj Gambhir. Hutter is a professor at the University of Freiburg with a decades-long career in automated machine learning; Hollmann was the lead author of the original TabPFN paper, which appeared in Nature in early 2025 and demonstrated that a transformer-based foundation model trained on synthetic tabular data could outperform gradient-boosted methods (XGBoost, LightGBM, CatBoost) on a wide class of real-world datasets without task-specific fine-tuning. The result was a landmark: for the first time, a single pretrained model could be downloaded and applied to arbitrary business data tables — sales figures, inventory records, customer churn datasets — without the weeks of feature engineering and hyperparameter search that traditional ML pipelines require.

For SAP, this is precisely the capability gap it needed to fill. SAP's S/4HANA and Business AI products sit on top of some of the world's largest enterprise data repositories, but turning that structured data into predictive intelligence has historically required custom ML projects that customers have neither the time nor the expertise to execute. TabPFN changes that calculus: a model that generalizes across tabular tasks without fine-tuning can be embedded directly into SAP's application layer, delivering ML-driven recommendations — demand forecasting, anomaly detection, procurement optimization — without a separate data science engagement.

SAP CEO Christian Klein, speaking at Sapphire 2026, confirmed that Prior Labs' research will anchor a new European Frontier AI Lab, positioned as an alternative to U.S.-headquartered AI research centers for enterprises that require GDPR-compliant AI development and on-premise deployment optionality. The combination also aligns with the European Commission's stated priority of building sovereign AI research capacity within the EU.

The NemoClaw Connection: Restricting Agent Access to Select Providers

SAP's announcement included one notable constraint for its AI agent ecosystem: customer-deployed AI agents within the SAP platform will be restricted to select approved providers, with Nvidia's NemoClaw explicitly named among those on the approved list. NemoClaw is Nvidia's enterprise AI agent orchestration framework, designed to run inference close to the data within customers' own infrastructure — compatible with SAP's on-premise deployment model and with European data residency requirements.

The restriction signals SAP's intent to control which inference providers can plug into its enterprise data layer. For SAP customers, this is a trade-off: they gain a curated, security-vetted set of AI agent capabilities, but lose the flexibility to swap in arbitrary open-source models or third-party agent frameworks without SAP's approval. Vendors not on the approved list — including some open-source frameworks with large developer communities — may find enterprise SAP deployments a closed market until they complete SAP's certification process. This dynamic is likely to generate pushback from the developer community and potentially from enterprise customers who want model choice freedom.

What to Watch

Three milestones define the practical impact of this deal. First, the approved provider list: SAP has named NemoClaw but has not published a full catalog of approved inference providers. The completeness and openness of that list will determine whether the acquisition accelerates European enterprise AI adoption or creates a proprietary lock-in that slows ecosystem growth. Second, TabPFN v3 benchmarks: Prior Labs has continued publishing research since the original Nature paper; the next major model update under SAP ownership will show whether the acquisition accelerates or constrains research velocity. Third, competitor response: Oracle and Microsoft have both announced competing enterprise AI data strategies in 2026 — Oracle through its OCI AI Infrastructure expansion and Microsoft through Copilot for M365. Watch whether either moves to acquire comparable tabular AI research teams in response to SAP's Prior Labs bet.

Want every major AI and Web3 funding round the moment it closes? Subscribe to the BlockAI News daily brief.