SAP Bets $1.16B on 18-Month-Old German AI Lab Prior Labs, Embraces NemoClaw Agents

Enterprise software giant SAP has announced a major investment in artificial intelligence, targeting the specific needs of corporate data management. On Monday, the European heavyweight declared its intention to acquire German AI startup Prior Labs for an undisclosed amount, with plans to invest €1 billion (approximately $1.16 billion) over the next four years. Pending regulatory approval, the acquisition will transform Prior Labs into a dedicated AI lab focused on structured data — the tables and databases that underpin most enterprise operations.

Key Facts Behind the Deal

• SAP will invest €1 billion ($1.16 billion) over four years into Prior Labs, which will operate as an independent AI lab.
• The acquisition price itself was not disclosed, but sources confirmed it was an "almost all cash" deal with well over half a billion dollars paid upfront to founders Frank Hutter, Noah Hollmann, and Sauraj Gambhir.
• Prior Labs was founded just 18 months ago in Freiburg, Germany, and focuses on tabular foundation models (TFMs) — AI models designed for data in rows and columns, unlike large language models (LLMs) that process text.
• SAP will maintain the open-source versions of Prior Labs’ TabPFN model series, which has been downloaded over three million times.
• Concurrently, SAP has updated its API policy to prohibit unauthorized AI agents, explicitly blocking OpenClaw but allowing NemoClaw agents through its endorsement of Nvidia’s Agent Toolkit.

Why Structured Data Matters for Enterprises

The investment underscores a growing recognition that enterprise AI needs differ from consumer-facing generative AI. While large language models like GPT-4 have captured headlines, they are often ill-suited for the structured data that powers accounting, HR, procurement, and expense management systems. Tabular foundation models are trained specifically on tables and databases, enabling predictions, classifications, and anomaly detection without the hallucination risks common in LLMs.

Prior Labs’ TabPFN family uses a novel architecture based on prior-data fitted networks, which can learn from small datasets and generalize across different table schemas. This is a critical advantage for enterprises that often have fragmented, siloed data. The models have been widely adopted in developer circles, with open-source downloads exceeding three million.

SAP’s Strategic Play in Agentic AI

The acquisition comes at a turbulent time for SAP. Its stock has dropped significantly in 2026 due to the "SaaSpocalypse" — a market correction affecting software-as-a-service companies. CFO Dominik Asam noted in January that the company must adopt new technologies quickly to maintain its economies of scale advantage. The Prior Labs deal is a direct response to the rise of agentic AI, where autonomous software agents perform complex tasks across business systems.

SAP is taking a defensive posture by restricting which agents can access its API. The new policy prohibits AI agents from entering through SAP’s API unless they use "SAP-endorsed architectures." This includes SAP’s own Joule Agents (still in beta) and agents built on Nvidia’s Agent Toolkit, which powers NemoClaw. Nvidia announced in March that SAP’s Joule platform supports its Agent Toolkit, giving NemoClaw — a security-focused variant of the open-source OpenClaw — a privileged position in the SAP ecosystem.

Previous AI Investments and Competition

SAP has not been idle in the AI space. In 2023, it backed OpenAI rival Anthropic, as well as European AI firms Aleph Alpha and Cohere (which now intend to merge). It also developed its own relational pretrained transformer model, SAP-RPT-1. However, SAP CTO Philipp Herzig acknowledged that the greatest untapped opportunity was in structured data. By acquiring Prior Labs, SAP gains immediate access to cutting-edge TFM technology and a team with strong academic roots.

The deal also contrasts with competitor Salesforce, which is taking a more permissive approach. Salesforce’s new Headless 360 architecture allows enterprises to choose their own agents, including OpenClaw. SAP’s walled-garden strategy aims to ensure security and compliance but risks alienating customers who prefer flexibility.

Founder Background and Future Plans

Prior Labs co-founders bring deep AI expertise. Frank Hutter is a professor of machine learning at the University of Freiburg and a leading researcher in automated machine learning (AutoML) and Bayesian optimization. Noah Hollmann and Sauraj Gambhir are PhDs with backgrounds in neural architecture search and probabilistic modeling. The trio raised $9.3 million in pre-seed funding from Balderton Capital in February 2025.

Balderton partner James Wise described the acquisition as "one of Germany’s biggest ever venture outcomes." The founders have promised to maintain open-source versions of their models, and the lab will operate independently to preserve research velocity. The ultimate goal, as stated by Hutter, is to build a "globally-leading frontier AI lab for structured data — in Europe, in the open."

Enterprise Implications and Market Reactions

Analysts see the move as a hedge against the commoditization of LLMs. As OpenAI COO admitted in early 2026, AI has not yet deeply penetrated enterprise business processes. SAP is betting that TFMs will bridge that gap by making AI truly useful for core business functions such as inventory management, fraud detection, and financial forecasting. The company will integrate Prior Labs’ technology into SAP AI Core and SAP Business Data Cloud, as well as its agentic layer Joule.

Following the announcement, SAP’s stock edged upward, reversing some recent losses. However, the broader SaaS market remains volatile. The Prior Labs acquisition signals that SAP is willing to make bold, long-term bets to secure its position in the next wave of enterprise AI.

Source: TechCrunch News