SAP’s AI Bet: Own The System Of Record, Rent Nobody’s Brain

SAP has launched Joule, an AI layer integrated into its core enterprise solutions, marking a strategic shift toward owning the data infrastructure that underpins AI capabilities. This move underscores SAP’s focus on controlling the data substrate rather than competing solely on model development, a decision that could reshape enterprise AI deployment.

As of mid-2026, SAP reports that Joule is active across more than 35 solutions, including S/4HANA Cloud, SuccessFactors, Ariba, and Datasphere, with over 30 specialized agents and 2,500 ‘Joule Skills.’ The company has committed a €100 million partner fund to enable system integrators to develop custom agents via Joule Studio, its low-code agent builder. SAP claims significant customer outcomes: a global retailer reducing HR process cycle times by 40–60%, an Argentine airport operator cutting costs by 16% and administrative effort by 90%, and developers experiencing approximately 20% productivity gains.

Strategically, SAP positions Joule within its ‘Autonomous Enterprise’ vision, where AI agents are considered as operationally significant as human users, joining the existing enterprise systems as ‘non-deterministic operators.’ The architecture relies on a Knowledge Graph that reads business metadata directly from SAP’s Business Technology Platform, ensuring contextually accurate responses tailored to specific workflows. This design emphasizes structured, permissioned data over open internet models, aiming to create a moat against competitors.

Furthermore, SAP adopts a model-agnostic approach, integrating third-party foundation models through recent acquisitions like Prior Labs. This allows Joule to orchestrate various models without dependency on a single provider, maintaining flexibility as the AI landscape evolves. The strategy also encourages customers to reduce custom code, aligning with SAP’s ongoing cloud migration efforts, and reinforcing a ‘clean core’ approach that simplifies AI integration.

Implications of SAP’s Data-Centric AI Approach

SAP’s emphasis on owning the data layer positions it uniquely in the enterprise AI landscape, where most competitors focus on developing or licensing models. This strategy leverages SAP’s vast installed base of mission-critical, regulated systems, giving it a competitive advantage in trust, compliance, and contextual understanding. However, reliance on third-party models and variable AI costs pose risks to predictable deployment and ROI. The move signals a potential shift in enterprise AI, from open model innovation to controlled, data-driven systems that prioritize reliability and governance.

SAP’s Historical Focus on Data and Enterprise Systems

For decades, SAP has dominated the enterprise software market with its ERP systems, handling transactions like purchase orders, invoices, payroll, and supply chain management for many Fortune 500 companies and the German Mittelstand. Its strategy has traditionally centered on providing reliable, compliant, and deeply integrated systems. The 2026 AI push, including Joule and the Knowledge Graph investments, builds on this foundation, aiming to extend SAP’s control into the AI-enabled automation of business processes. This approach contrasts with the broader industry trend of frontier labs developing large models, instead focusing on owning the structured data that underpins enterprise operations.

“SAP’s move to own the data infrastructure rather than chase model IQ fundamentally shifts the enterprise AI landscape, leveraging its existing data moat.”

— Thorsten Meyer, AI Strategy Expert

Uncertainties in Adoption and Cost Management

It remains unclear how widely SAP’s Joule will be adopted across diverse enterprise environments, especially given the variable costs associated with consumption-based AI features. The €100 million partner fund aims to subsidize demand, but actual customer engagement and ROI are still unverified. Additionally, reliance on third-party models introduces dependency risks if model quality or access changes unexpectedly.

Next Steps for SAP’s AI Ecosystem Expansion

SAP will likely focus on scaling Joule’s deployment, expanding its agent ecosystem, and refining cost models to encourage broader adoption. Monitoring customer case studies and feedback in the coming quarters will be critical to assess whether the strategy delivers measurable ROI and sustains enterprise trust. Further investments in the Knowledge Graph and model orchestration capabilities are expected to reinforce SAP’s position as the data and AI infrastructure provider for large enterprises.

Key Questions

How does SAP’s Joule differ from other enterprise AI solutions?

Joule is integrated directly into SAP’s core enterprise systems, emphasizing ownership of structured, permissioned business data and orchestrating third-party models, rather than relying solely on open internet models or building proprietary AI models.

What are the main risks associated with SAP’s AI approach?

The key risks include unpredictable AI costs due to consumption pricing, dependency on third-party models whose availability and quality may change, and slow adoption driven by enterprise compliance, trust, and ROI concerns.

Why is SAP focusing on data ownership rather than model development?

Owning the data layer provides a competitive moat, leveraging SAP’s existing enterprise data infrastructure, and ensures AI applications are trustworthy, compliant, and contextually accurate, which is critical for mission-critical business processes.

Will SAP’s strategy limit innovation compared to frontier labs?

While it may limit the rapid development of new models, SAP’s approach prioritizes stability, compliance, and control, which are essential for large-scale enterprise deployment. Its model-agnostic orchestration allows flexibility and adaptation as the AI landscape evolves.

Source: ThorstenMeyerAI.com