Presented by SAP
In an era where anyone can spin an LLM, the real differentiator is not the AI technology itself, but the institutional knowledge base. Internal and partner consultants lead operational change when their recommendations impact integrated processes across supply chain, manufacturing, finance and other core functions.
"Grounded AI is non-negotiable, because accuracy is not optional when we are doing million-dollar transformation projects within the SAP ecosystem, for example," says Natalie Hahn, VP and Chief Product Officer, General AI at SAP Business AI. "Generation technology beyond retrieval, and the ability to anchor responses in trusted enterprise knowledge, helps ensure accurate code interpretation, best practice guidance, and clean decision support. This is how we bring real confidence to AI-powered consulting."
He added that there is tremendous value in product use cases for fully grounded AI assistants like SAP Jewel for advisors. SAP Joule has terabytes of enterprise data that is constantly generated and updated, so a consultant can be assured that they are getting up-to-the-minute SAP best practices and procedures when relying on Joule, while at the same time accelerating project delivery.
"We’re saving 14% of rework time, and saving advisors 1.5 hours per user per day, which is huge when you consider how expensive advisors are now." Han says. "Early adopters such as Wipro estimate that they have saved 7 million hours on a manual basis for their consultants."
Foundation of SAP Joule
Sachin Kora, chief architect of SAP Business AI, says that SAP Jewel is as certified as any consultant. The tool was born in 2023, when GPTS famously passed a simulated bar exam and there was a flurry around LLM’s ability to handle large amounts of context. It is widely recognized that the SAP ecosystem, with its associated domain ontologies and hierarchies, is incredibly vast and can be very complex to navigate. The question became how could an AI co-pilot be used to navigate this complexity when it was actually dropped within the SAP ecosystem itself?
Sachin Kora started experimenting with Frontier LLM models through the same certification exams. Early results were poor, but after extensive contextual tuning and a focus on delivering value to the partner ecosystem, Joule now consistently scores 95 percent or higher.
"Not only were we testing from a data point of view, but we were able to work with all of our advisors to create what we call the Golden Data Set," Han added. "It is unstructured, language-based, and well-grounded in human consultant expertise. We partnered with the entire consulting organization to manually label all products with Golden Data. It still forms the basis of everything."
State-of-the-art indexing pipeline
Joule for Advisors stays updated in real time. A sophisticated indexing pipeline pushes new SAP documents and releases content in the model as it is published, giving consultants confidence that each response reflects the most current guidance.
"This is pure engineering work done by our data scientists and engineers using many core SAP technologies," Cora explains. "We leverage the SAP business foundation layer, document grounding services, and many purpose-built systems to stay on top of current events in the system."
SAP Business AI also has board-level alignment, ensuring that it is not just a team effort but a company-wide priority. He has built strong internal partnerships with content owners across SAP – including SAP Learning, SAP Community, SAP Help, product teams, and consultant teams. Together, they continuously update proprietary content such as SAP Notes, Knowledge Base Articles (KBAS), and other domain-specific guidance that reflect evolving SAP best practices.
All this means for advisors Joule can take this constantly updated data and provide answers in near real time. This is the kind of research that would otherwise take consultant hours. But information pulled directly from the source makes available current and authoritative guidance to consultants, helping to eliminate early-stage misses that can derail a project months later when scoping wasn’t aligned with the latest capabilities.
Ensuring enterprise-grade security
Hahn says SAP is building a product that is relevant, reliable and responsive. As a company based in Europe, it takes data privacy seriously, following GDPR and other EU company regulations. At the core of SAP Business AI is AI Foundation, an AI operating system that governs AI with built-in security, ethics, and orchestration, using automation and intelligence to manage lifecycles, optimize resources, and promote flexibility.
All LLMS SAP and its users work within the AI Foundation, which protects private and proprietary data from leakage. Beyond data protection, SAP treats privacy, ethics and security at an enterprise level, with humans in the loop to run checks and balances.
"We have an enterprise-grade security framework as well as rapid injection and guardrail testing," Korra says. "An orchestration layer built within AI Foundation anonymizes inputs as well as moderates them to prevent malicious content. This ensures that the output we give to our customers is relevant to the SAP ecosystem, to the domain they are asking about, and not just a generic LLM. This set of tools, from the framework layer to the application layer to product quality, and also the most thorough testing necessary to secure our products. Then and only then it can reach our customers and partners."
Pushing the boundaries of joule for counselors
"We are barely scratching the surface of what LLMS and Agentic AI have to offer," Han says. "Access to knowledge is only the beginning. We are going to have a deeper understanding of the customer’s SAP system and be able to help them implement and transform their journey. The product team and our engineers are working to further transform the tool, be able to detect more insights, connect to customers’ systems, and understand and improve their processes, including developing code and handling customer code migration."
The next step is adding the second layer of grounding. SAP’s customer base is vast, and its partner ecosystem has implemented countless business scenarios. Grounding Jolly in SAP’s institutional knowledge was the first milestone. Next is laying out each user’s own proprietary context—historical system data, process designs, implementation blueprints, and internal documentation. This makes Joule aware of the SAP wire to the customer, and provides guidance that is in line with how the business actually operates.
“Think of it as layering your knowledge on top of SAP’s knowledge – giving you more accurate and relevant guidance,” says Cora. “Information that might otherwise be lost can sit on top of a jug for advisors. Our system processes it and makes sure it gets to you in the right way and at the right time.”
This expanded grounding also allows Jewell to adjust his mentorship to a consultant role—whether serving as an architect, functional consultant, or technical advisor.
"We provide the information they need to create a particular customer," Han explains. "Then we can answer not only the general questions, but we can answer them in their particular order. From there it’s a step forward in generating more insight and taking further action."
Sponsored articles are content produced by a company that is either paying for the post or has a business relationship with VentureBeat, and is always clearly marked. For more information, contact sales@ventorbet.com.