User interfaces are key. As the primary conduit and touchpoint for users (and, at some level, machine-to-machine interfacing), the Graphical User Interface (GUI) has long been established as our central focus for PC and mobile device interaction. While that standard isn’t likely to change in fundamental terms, the rise of ‘copilot’ technologies driven by generative AI is bidding to skew (and improve) the way we interact with applications and data services.
German softwarehaus SAP thinks it has seen this trend play out and last year announced the release of its own generative AI assistant. Known as Joule and named after the 19th-century English physicist James Prescott Joule, this natural language technology is now undergoing extensive integration throughout the company’s technology platform and portfolio of business applications.
Bolstered Human Workflows
Capable of being rather more than ‘just’ a chatbot, SAP says that Joule is capable of sorting and contextualizing data from multiple systems to surface business insights in a in a secure and compliant way. As human workflows are now bolstered (and perhaps emboldened) by our new smart assistants in this way, the key interest point now will come down to just how far, wide and deep we allow these technologies to penetrate the normal course of business life.
In SAP terms, the penetration remit can be as far-reaching as possible.
Joule was initially tabled for integrations in SAP SuccessFactors and SAP Start – these being SAP’s Human & Capital Resources (HCM) and the company’s out-of-the-box default central entry point ‘home page experience’ running on the SAP Business Technology Platform respectively.
Since its September 2023 initial integrations, SAP has integrated Joule into SAP S/4HANA Cloud Public Edition and SAP S/4HANA Cloud Private Edition, as well as in multiple products including SAP Customer Data Platform, SAP BTP Cockpit, SAP Build, SAP Build Code and SAP Integration Suite. In the second half of 2024, Joule is also planned to be available in SAP Ariba, SAP Analytics Cloud and in multiple supply chain management solutions from SAP.
¿Hablo Generative AI?
Additionally, SAP notes that Joule will now be able to understand and respond to queries in a growing number of languages, including German, Spanish, French and Portuguese.
In terms of actual practical use cases here, employees can use Joule to get answers to natural language questions about their organization’s HR policy documents in SAP SuccessFactors. The Joule copilot also provides a conversational way for workers to ask questions and complete HR processes.
“These benefits are enabled by Joule’s new ‘document grounding’ capability, which can provide more comprehensive responses by drawing from business documents located in SAP and third-party repositories such as Microsoft SharePoint. This capability deepens how Joule reliably delivers answers grounded in customers’ structured and unstructured business data,” noted SAP, in a technical product statement.
The company explains how Joule is now being feathered out across its stack and says that commodities traders (for example) will be spared from repetitive data entry tasks in the SAP S/4HANA Cloud Private Edition using its commodity management for physical contracts solution. With Joule, business operators can create new commodities deals by simply describing the deal using conversational language, giving them more time to focus on producing better commercial outcomes for their clients. This capability is planned to be available in the second half of this year.
Bejeweled by Joule, Everywhere
As SAP now works to extend its copilot assistant throughout its stack, we can expect to see it aligned for use cases handled by everyone from billing specialists to procurement professionals to project managers and onwards. Business analysts and process owners can also automate business processes at scale using generative AI capabilities made possible by Joule.
What comes next? Logically we might expect to see software application developers get more code completion, code testing and code compliance support through this technology at a lower base level of functionality. Then, perhaps, outward extension from the SAP motherlode to partner technologies as well as deeper integration with key generative AI leaders already showing their hand such as Microsoft Copilot (where SAP has already cemented a partnership) and beyond.
