Kennisbank
Why the promise of AI is not always fulfilled
- By Nidaros
AI takes over work, but often gives no insight in return. The result: employees who remain responsible for processes they no longer understand.
More and more organizations are using Robotic Process Automation (RPA) and Artificial Intelligence (AI) to automate routine tasks and administrative work. Software robots execute repetitive processes accurately. AI adds intelligence to this: it recognizes patterns and makes decisions based on data.
The benefits are obvious. For organizations, it means higher productivity and lower costs. Employees benefit because monotonous work disappears, creating space for more creative and strategic tasks with greater autonomy and job satisfaction.
Opportunities and risks
However, that promise is not always fulfilled. Alongside opportunities, there are also risks – especially when RPA and AI are implemented without a clear strategy. The work that remains often becomes more complex and demanding, with less room for recovery time. Or worse: employees remain responsible for processes they no longer understand, because AI has taken over the work without providing transparency.
Gerben Dolsma recognizes this problem. He is founder and CEO of Nidaros in Hoogeveen, a company that helps organizations become more productive and efficient by automating repetitive actions and processes.
“A virtual robot uses existing software applications and systems, just as an employee would,” he explains. “In the background, the robot takes over time-consuming, repetitive tasks.”
Insight and control
Dolsma regularly sees organizations struggle when they focus only on efficiency.
“You shouldn’t just give people time back – you must also give them insight and control,” he states. “Otherwise, you become efficient and powerless at the same time.”
An example: a customer service employee receives a call about an invoice. Previously, she had processed it herself and could immediately help. Now AI has taken over the processing. When the customer asks questions, she no longer knows how the invoice was created or why certain amounts were calculated. The expert has become little more than a relay point.
The problem is not automation itself. It lies in how automation is implemented.
“Most employees are deeply involved in their organization and want to help customers as well as possible,” Dolsma explains. “They feel responsible for the outcome, but often do not understand what AI is doing behind the scenes. As a result, they experience no influence. That imbalance between high engagement and low influence creates stress and can ultimately lead to burnout.”
The one-minute rule
Nidaros therefore follows a simple principle: do not remove people from the process – place them above it. In what could be called a control room, employees see in real time what is going right and what is going wrong. They manage exceptions, not daily routine. In this way, you retain efficiency while giving people control.
“In our approach, we apply the one-minute rule,” Dolsma clarifies. “If a customer calls, an employee must be able to explain within one minute what is happening. Not because they execute every process themselves, but because they have full transparency. They see when something came in, why certain decisions were made, and where something may have gone wrong. And in case of an exception, they can see with one click: is this a technical error, a functional rule issue, or a business question requiring human judgment?”
Cultural change
RPA and AI work – but only if implemented correctly. Many organizations believe it is about technology, while at its core it is about people. In other words, it is not a technical issue, but a strategic one.
“It is a cultural shift,” Dolsma emphasizes. “No human colleagues are added – but virtual ones are. And you must learn how to collaborate with them. That is fundamentally different from implementing a new IT system. Just like with human colleagues, you must understand each other and build trust. Otherwise, it becomes a black box, with all the consequences that entails. That is why insight and control are not nice-to-haves, but prerequisites for real efficiency and job satisfaction.”
Tags:
- AI, Automation, Culture, Process Design, RPA, Team
