A tech confession and a debate with an AI robot was the recipe for the future of work at SAP Connect Southern Africa in Kyalami near Johannesburg last week.
“I have a confession to make: I have been afraid of AI,” said Stanley Dube, SAP head of solution advisory. Standing alongside a robot wearing a tracksuit carrying SAP’s AI assistant brand Joule, Dube said: “The fear I have had is that AI is going to take my job.”
However, having moved past that fear, he and Joule addressed questions about how organisations can adapt to AI, redesign jobs and processes, maintain governance and build a competitive advantage. In a theoretical example, Dube asked about the specifics of a project.
“I already know the status,” answered Joule. “I also know who is usually responsible for the delays: you, Stanley Dube.”
Debate around AI replacing workers and robots taking control has become familiar, but SAP’s demonstration gave Joule a physical presence beyond a computer screen. The robot appeared to serve as a representation of the software assistant rather than a demonstration of autonomous physical AI. Although the presentation carried an element of theatre and Joule’s responses may have been preprogrammed, the exchange pointed towards a future in which AI agents could take on a more visible and collaborative role in the workplace.
Dube asked Joule: “What do we, as entities and companies, actually have to do to survive and thrive in the age of AI?”
Joule answered: “Most organisations do not have an AI problem; they have an adaptability problem. [This is] a massive waste of time and capital. I am built directly into your SAP applications. While you slept, I autonomously reconciled 84% of your finance accounts, predicting a pump failure at Secunda to generate a maintenance order and rerouted logistics around a major issue.”
Joule’s response framed the main obstacle as organisational rather than technological. AI systems may process information and act within seconds, but the benefit can be lost when decisions still depend on lengthy approval chains and slow human processes. In that situation, advanced technology becomes the equivalent of placing a Ferrari engine in a donkey cart: the system has considerable power, but the surrounding organisation cannot move at the same speed.
Dube said employees might find themselves arguing with AI agents in future, but that should not necessarily be a cause for concern.
Joule said: “Remember, AI is the enabler. Adaptability is the challenge. The future belongs to those who learn how to adapt because of AI, not just those who adopt.”
The presentation extended a strategy outlined at SAP’s Sapphire conference in May. As Gadget previously reported, SAP presented the Business AI Platform as an environment for developing and governing agents that work across enterprise operations. The approach uses ERP systems as the source of business data, processes and controls, allowing agents to perform tasks within established organisational boundaries.
Standard Bank provided a practical example of the preparation required to support SAP’s vision. The bank has modernised core enterprise systems using SAP technology, establishing a platform from which further automation and AI applications could develop.
Vanessa Padiachee, Standard Bank Group CIO for corporate functions, said: “If you look at the pace of change, technology is changing rapidly. AI is no longer just around experimentation and production use cases. Organisations are now going beyond that to look at enterprise scaling use cases.”
Padiachee said SAP was an important part of Standard Bank’s broader technology programme. She said organisations must manage rapid technological change while keeping major systems aligned with practical business requirements.
Standard Bank is focused on modernising the systems needed to support future automation and AI deployments rather than describing an autonomous AI implementation.
That work receives far less attention than demonstrations of agents and automated decisions. However, SAP’s autonomous-enterprise vision still depends on businesses having systems, processes and employees capable of supporting the technology.
Joule may have stepped beyond the computer screen at SAP Connect, but the route towards an autonomous enterprise still runs through the less theatrical work of modernising systems, data and processes.
* Jason Bannier is a data analyst at World Wide Worx and deputy editor of Gadget.co.za. Follow him on Bluesky at @jas2bann.