When the robot arrives, but nobody wants it

A humanoid robot stands in the factory floor. Employees either pull out their cell phones or step back. The sales presentation on cycle times and ROI continues in the background—but the real question hangs unanswered in the room: Will this replace me?

This shows that AI change management often falls short in practice. Common frameworks were developed for software rollouts: a new ERP system, a chatbot in customer service, an AI-powered reporting tool. But as soon as AI takes the form of a robot on the factory floor, the situation changes fundamentally.

This becomes particularly evident where robotics is integrated directly into the SAP-driven process chain. Many S/4 conversion projects in production and logistics today go hand in hand with the use of robotics on the shop floor: MES systems synchronize the manufacturing steps, EWM controls material flows—and robotics handles the physical execution. For employees, this means that the robot is not a foreign element alongside the SAP system, but rather a visible component of a process that was previously mediated through screens and menus.

What Makes the Difference

An Unmissable Presence: A software tool can be ignored or left open in a browser tab. A robot next to the workbench requires a constant physical presence.

Concrete job anxiety: When software is rolled out, we often talk about a shift in responsibilities—new tools, new processes—but the role remains essentially the same. With robotics in production, the logic of substitution is immediately apparent: “This movement—the machine can do that now, too.” The level of abstraction, which still leaves room for reassurance when it comes to fears about software, is absent here.

Real safety concerns: With a new CRM system, we’re talking about data security. With a robot moving in the same physical space as people, we’re talking about operational safety, CE compliance, emergency shutdown procedures, and liability issues. Employees intuitively sense this difference, even without knowing the relevant standards.

A Practical Framework for Physical AI

Based on project experience, we can identify five levers that effectively complement traditional change management approaches:

1. Prioritizing Acceptance According to the Kano Model: This model lends itself perfectly to acceptance efforts: Which functions are basic requirements, which are performance features (the better they are, the greater the satisfaction), and which are features that generate enthusiasm? In robotics, it’s often the case that safety functions are basic requirements—here, complete trust is essential. Enthusiasm arises, for example, when the robot takes over physically taxing tasks and directly reduces the user’s workload.
2. Early Involvement of Employee Representatives: Works councils have extensive rights of co-determination when automation technology is introduced and convey the workforce’s concerns unfiltered. Delayed communication erodes the trust that is so essential.
3. Robotics-as-a-Service as a Psychological Lever: When a robot is marketed as a one-time capital investment, the implicit message is that it is a „replacement.“ A service model (similar to recurring maintenance contracts at an auto repair shop) shifts the perspective toward a system that receives ongoing support and requires human supervision.
4. Incorporating Technical Questions into Change Communication: CE certification, service-level agreements, and response times in the event of malfunctions should not only be included in the specifications but also communicated to the workforce in order to alleviate abstract fears.
5. Using the SAP Interface as a Trust Anchor: When robotics is controlled via familiar SAP interfaces and error messages appear in the familiar system, the sense of unfamiliarity diminishes. The machine no longer seems like an unpredictable black box, but rather as an executing component of a familiar tool.

What Matters Now

Robotics is still a niche topic in manufacturing, but falling costs for humanoid autonomous systems will soon lead to the widespread adoption of physical AI—for example, in logistics or healthcare.

Change managers must learn that the obstacle does not lie in the technology itself, but in the immediacy with which it is experienced. Those who understand this can prevent transformations from failing due to a lack of trust among the workforce.