Why did SAP consultants reject AI work initially?

Consultants initially rejected AI-generated work due to preconceived bias, only recognizing its accuracy when they believed it was produced by human interns.

How does SAP's AI co-pilot Joule help consultants?

Joule for Consultants removes clerical work, making consultants more effective and allowing them to focus on business insights rather than technical systems.

What is SAP's vision for the consultant of 2030?

SAP envisions consultants as deeply human practitioners, enabled by AI, focusing on strategic thinking and customer goals rather than technical execution.

Home / Technology / Human Bias vs. AI Accuracy: SAP's Shocking Test

Human Bias vs. AI Accuracy: SAP's Shocking Test

10 Dec

•

Summary

Consultants rated AI work 95% accurate when thought to be from interns.
The same AI work was rejected when identified as AI-generated.
SAP aims for consultants to be AI-enabled, not replaced.

Human Bias vs. AI Accuracy: SAP's Shocking Test

An internal SAP experiment has exposed a significant bias among consultants regarding artificial intelligence. When consultants believed AI-generated business requirement analyses were completed by junior interns, they rated the work 95% accurate. However, upon learning the same analysis was produced by SAP's AI co-pilot, Joule, they dismissed nearly all of it, only recognizing its accuracy upon individual review.

This finding underscores SAP's caution in introducing AI, particularly in how senior consultants perceive its capabilities and integration into workflows. The company emphasizes that AI copilots like Joule are designed to augment, not replace, expertise. By handling routine clerical tasks, AI frees up consultants to concentrate on delivering high-quality answers more effectively and rapidly, shifting their focus from technical execution to business insight.