SAP Experiment Exposes 'Algorithmic Aversion': Veteran Consultants Show Bias Against AI-Generated Work

2 days ago7 min read

A revealing internal experiment at SAP has provided a real-world case study in 'algorithmic aversion,' demonstrating a powerful human bias that could influence how artificial intelligence is adopted across the corporate world. The study tasked five teams of seasoned SAP consultants with reviewing over a thousand business requirements addressed by the company's AI assistant, Joule for Consultants.The outcome highlighted a critical trust gap: four teams, told the work was completed by junior interns, rated its accuracy at an impressive 95%. The fifth team, informed the identical output came from an AI, rejected nearly all of it immediately.This blanket dismissal only subsided when consultants were required to evaluate each answer individually, discovering the AI's analysis was, in fact, highly accurate and detailed. According to Guillermo B.Vazquez Mendez, SAP America's chief architect for RI business transformation, the lesson is clear: how AI is communicated and integrated is paramount, especially when introducing it to experienced professionals whose deep expertise can breed skepticism. This is more than a test of a tool's efficiency; it's a foundational story about trust and the undervaluing of machine-generated work.The resistance is understandable. Consultants with decades of experience possess tacit knowledge that is difficult to codify.SAP's approach positions Joule as an amplifier of human capability, not a replacement. By handling the technical groundwork—which historically consumed 80% of a consultant's time—the AI rebalances the equation.It allows consultants to shift from system mechanics to business strategy, applying their irreplaceable human judgment to higher-value outcomes. The experiment also revealed a generational adoption pattern.Joule serves as a bridge, enabling new hires to ramp up quickly and work more independently, while senior consultants engage with it for complex, insight-driven tasks. Observing juniors' success with the tool can help overcome initial bias among veterans.Looking ahead, the current phase of AI, which requires skilled prompting, is just the beginning. The future points toward 'agentic' AI systems that can interpret entire business processes and act autonomously.SAP, with its repository of over 3,500 mapped business processes underpinning global commerce, is uniquely positioned to develop such agents. The journey from biased rejection to trusted collaboration will ultimately determine whether AI becomes a true partner in progress or remains hampered by unexamined prejudice.

#featured

#SAP

#Joule for Consultants

#AI skepticism

#prompt engineering

#business transformation

#AI adoption

#human-AI collaboration