Skip to main content

Data & AI

SAPPHIRE 2026: OUR EXPERT KEY TAKEAWAYS 

Published on 05/20/2026

SAP Sapphire 2026 had one very clear message: enterprise software is about to become a lot more helpful, and a lot less frustrating 

So far, the usage of AI credits among SAP customers has remained relatively low, which may suggest that SAP has not yet fully convinced customers of the practical value of its AI capabilities. 

Sapphire 2026 clearly showed how SAP intends to change that, with a wave of innovations designed to make AI more useful, contextual, and operationally relevant. 

Let’s start with the concept of the Autonomous Enterprise, where AI agents and human teams work together to execute business processes faster, smarter, and with fewer endless handoffs. 

The SAP Innovation News Guide is packed with new Joule Assistants and agents for areas such as financial closing, billing, sourcing, logistics, treasury, and workforce planning. I strongly recommend reading it: 

The overall direction is easy to spot: SAP wants AI to handle repetitive work, identify issues earlier, and reduce the number of times users need to jump between systems simply to complete one task.

In other words: less “process theatre,” more actual process.And who is doing the work (I should say “what”, let’s be welcoming to our new AI colleagues.): Joule Agents.

Forget the idea of SAP Joule as a chatbot that merely triggered predefined actions through “skills.”

One of the most important updates announced at Sapphire 2026 is Joule Work, a dynamic workspace where users can describe what they want in natural language while Joule Assistants coordinate execution across SAP and non-SAP applications.

Joule Work becomes a real-time operational interface where users can launch processes, validate outputs, supervise execution, and collaborate directly with AI assistants. And yes, if needed, you can still launch classic Fiori tiles.

Joule Work also introduces a new model of collaboration between humans and AI. Joule Assistants act as intelligent coordinators, while Joule Agents execute specific operational tasks across systems, processes, and data sources. Think of Joule Assistants as the managers of your Joule Agents.

SAP also possesses something uniquely valuable for AI: decades of enterprise application data. SAP processes generate enormous amounts of highly structured business information stored in complex ERP tables, transactional histories, approvals, workflows, and operational relationships.

SAP is now finding clever ways to unlock this treasure trove through technologies such as Domain Models, Knowledge Graphs, and Business Data Cloud. This allows SAP AI to become significantly more intelligent, contextual, and operationally relevant than generic AI solutions.

Domain models are domain-specific models trained on SAP knowledge including tables, documents, processes, and enterprise relationships.

They help Joule Agents and Joule Assistants because they provide the business context AI needs to behave as if it truly understands the company, not just the words written in the prompt. 

SAP’s domain models structure enterprise knowledge in a way that allows AI to understand relationships between invoices, purchase orders, approvals, supply constraints, and financial close activities. This helps Joule avoid generic answers and instead generate actions that align with SAP workflows, controls, and business guardrails.

It also improves scalability, because SAP can reuse the same business semantics across multiple assistants and agents in finance, procurement, supply chain, HR, and beyond.

SAP also introduced the concept of “Company Memory” which aims to integrate enterprise documentation, emails, and internal knowledge into AI workflows to make agents even more effective.

All these innovations are relying on an advanced platform that combines all the required elements, and is comprised of elements from SAP BTP, SAP BDC and others, the SAP Business AI Platform.

The headline this year is the SAP Business AI Platform, which SAP says will anchor its new AI strategy. The objective is not simply to add AI on top of enterprise software, but to make AI part of the business operating model itself — with enough context to understand enterprise processes, controls, and data.

That might sound lofty, but the core idea is actually very practical: Business AI needs to understand the difference between “move fast” and “oops, that invoice was wrong.” SAP is leaning into that distinction by combining AI with its ERP and process expertise. 

CONCLUSION

SAP is trying to transform enterprise software from a place where work gets documented into a place where work gets done. If SAP succeeds, the future of enterprise software may involve far fewer screens, clicks, and process bottlenecks, and far more collaboration between human teams and intelligent agents. 

The future experience of enterprise software may simply become:

“Tell the system what you need — and let it help carry the load.”