SAP’s $1.16 Billion AI Bet: What NemoClaw Means for This Young German Lab

Key Takeaways:

1. Strategic AI Acquisition: SAP is making a significant push into enterprise AI with the acquisition of German startup Prior Labs and a €1 billion investment over four years, aiming to create a leading AI lab for structured data.
2. Focus on Tabular Foundation Models: Unlike many in the AI race, SAP is prioritizing Tabular Foundation Models (TFMs) designed for the structured data prevalent in enterprise systems, positioning Prior Labs as a crucial asset for this specialized niche.
3. Controlled Agentic AI Ecosystem: SAP is adopting a strict, “walled garden” approach to AI agents, explicitly prohibiting unauthorized agents from accessing its APIs, while endorsing its own Joule Agents and select partners like Nvidia’s NemoClaw, contrasting sharply with more open competitor strategies.

The enterprise software landscape is undergoing a seismic shift. As OpenAI COO’s own admission last February highlighted, “we have not yet really seen AI penetrate enterprise business processes.” This lack of deep integration, coupled with the ongoing “SaaSpocalypse” — a period of re-evaluation and pressure on Software-as-a-Service valuations — has left major players like European heavyweight SAP facing significant challenges, reflected in a notable drop in its stock value in 2026. However, far from retreating, SAP is doubling down, strategically positioning itself to not just survive but thrive in the age of artificial intelligence.

SAP’s Bold Bet: Acquiring Prior Labs for Structured Data AI

In a move poised to redefine its AI strategy, SAP announced its intention on Monday to acquire German AI startup Prior Labs for an undisclosed sum. This acquisition is not merely a purchase; it’s a foundational investment. Pending regulatory approval, SAP plans to inject a staggering €1 billion (approximately $1.16 billion) into the nascent business over the next four years. The ambitious goal: to transform Prior Labs into a world-class AI lab specifically dedicated to structured data – the vast, organized tables and databases where the lifeblood of enterprise information typically resides.

While the exact acquisition cost remains under wraps, sources close to the deal informed Pathfounders that it represents an exceptionally “healthy exit” for Prior Labs’ founders. The deal involved a substantial “almost all cash” payout, with well over half a billion dollars in cash upfront distributed among the startup’s co-founders: Frank Hutter, Noah Hollmann, and Sauraj Gambhir. This lucrative outcome for a company barely 18 months old underscores the immense value SAP places on its specialized technology.

Founded just a year and a half ago, Prior Labs quickly gained traction for its pioneering work on tabular foundation models (TFMs). These specialized AI models are engineered to make highly accurate predictions and derive insights directly from data organized in tables and databases. This focus represents a critical differentiator from the more widely publicized large language models (LLMs), which excel with unstructured text. For enterprise giants like SAP, whose widely used software products for accounting, HR, procurement, and expense management are inextricably linked to vast internal databases, TFMs offer a potentially superior and more precise fit for their core business needs. TFMs inherently understand the relationships and constraints within structured data, leading to fewer hallucinations and more reliable outputs critical for business operations.

Beyond Acquisition: SAP’s Broader AI Vision

This acquisition isn’t SAP’s first foray into the AI space. The German company has been actively investing in generative AI companies, spanning both large and small language models. In 2023, it strategically backed OpenAI rival Anthropic, alongside Aleph Alpha and Cohere, the latter two now famously intending to merge into what they envision as a “global AI powerhouse.” Internally, SAP had also developed its own relational pretrained transformer model, SAP-RPT-1.

SAP CTO Philipp Herzig articulated the strategic rationale behind prioritizing structured data AI: “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.” The acquisition of Prior Labs provides a significant shortcut in realizing this vision, offering proven technology and a talented team. Prior Labs’ TabPFN model series has already seen substantial developer traction, with its open-source models downloaded over three million times, as proudly announced by the founders in a blog post about the deal.

In its press release, SAP pledged that Prior Labs would maintain its commitment to open source, ensuring continued access to its popular models. The new lab will operate as an independent unit, designed to foster rapid research and development while leveraging SAP’s massive resources for long-term investment and a direct pathway to productization across the entire SAP portfolio. This integration will occur through platforms like SAP AI Core and SAP Business Data Cloud, as well as the crucial agentic layer provided by Joule. The hope is that Prior Labs, with this “massive boost” from SAP, will evolve into a new “globally-leading frontier AI lab for structured data — in Europe, in the open,” a sentiment echoed by founder and CEO Frank Hutter on X.

Navigating the Agentic AI Frontier: Defense and Control

Beyond its focus on structured data, Germany’s most valuable company is also acutely aware of the burgeoning trend toward agentic AI – intelligent agents capable of autonomously executing tasks. In this arena, SAP appears to be playing a defensive game, opting for a highly controlled ecosystem. As The Information first reported, SAP has moved to block OpenClaw and any other agent technology that it has not explicitly authorized.

When queried for comment, SAP’s press department directed TechCrunch to the company’s latest API policy. This policy explicitly states that SAP “prohibits” AI agents from accessing its products through its API, with the sole exception of those operating within “SAP-endorsed architectures.” Naturally, these authorized architectures include SAP’s own offering, Joule Agents, which is currently in beta and allows customers to create their own AI agents for specific tasks within the SAP ecosystem.

Adding another layer to its strategy, Nvidia announced in March that SAP’s Joule supports Nvidia’s Agent Toolkit, a software framework for managing agents. This toolkit forms the foundation for Nvidia’s enterprise-ready, security-focused OpenClaw competitor, NemoClaw. Consequently, SAP customers will be authorized to use NemoClaw agents, indicating a strategic partnership rather than an open-door policy. This approach allows SAP to maintain tight control over security, data integrity, and performance, crucial aspects for enterprise-grade applications.

For a giant incumbent like SAP, AI presents a dual-edged sword: both an existential threat if ignored and an unparalleled opportunity if harnessed effectively. As CFO Dominik Asam articulated to CNBC in January, “It’s all about how quickly [we can] as SAP actually also embark [on] these technologies in our R&D portfolio to keep the relative economies of scale advantage.” The strategic investments and controlled agentic framework are direct manifestations of this philosophy.

Market Reactions and Contrasting Strategies

Prior Labs, headquartered in Freiburg, Germany, had previously raised approximately $9.3 million in a pre-seed funding round in February 2025, led by Balderton Capital. While this was more than competitor Neuralk-AI had secured, it paled in comparison to the $255 million Series A raised by Fundamental upon its emergence from stealth in the same month. Despite the smaller initial funding, Balderton partner James Wise lauded Prior Labs’ acquisition on X as “one of Germany’s biggest ever venture outcomes,” validating the startup’s rapid growth and innovative technology. Following the announcement, SAP’s stock saw a slight upward trend, indicating positive investor sentiment toward its proactive AI strategy.

SAP’s highly restrictive stance on AI agents stands in stark contrast to the approach of other incumbents grappling with the “SaaSpocalypse.” Salesforce, for instance, has embraced a more open philosophy. With its new Headless 360 architecture, Salesforce is allowing enterprises to choose their own agents, including OpenClaw if they wish, emphasizing customer choice and flexibility. This divergence highlights a key strategic debate within the enterprise software industry: whether to control the AI ecosystem tightly for security and integration, or to open it up for broader innovation and customer freedom. SAP’s decision underscores its conviction that a controlled, deeply integrated AI future, especially for mission-critical structured data, is the path to long-term stability and competitive advantage.

Bottom Line

SAP’s acquisition of Prior Labs and its significant investment in structured data AI, coupled with a deliberate, controlled strategy for agentic AI, marks a decisive turning point for the enterprise software giant. Faced with market pressures and the imperative to deeply integrate AI, SAP is not merely experimenting; it’s building a foundational pillar for its future. By focusing on tabular foundation models and maintaining a tight grip on its agent ecosystem, SAP aims to leverage its existing strength in structured data to deliver reliable, enterprise-grade AI solutions. This bold move positions SAP to not only address the challenges of the “SaaSpocalypse” but to potentially emerge as a dominant force in a specialized, yet crucial, segment of the evolving artificial intelligence landscape, setting a distinct course from its competitors in the race for AI supremacy.