AI, AI, AI. If you only put people “in the loop” of AI that no one fully understands anymore, you lose control. Turn it around: as a team, be the architect of your business processes and put AI in the loop, within your purpose, your goals and the interests of all stakeholders.
Sound familiar? Everywhere you look, opportunism and fear seem to be in charge. AI, AI, AI. “This is an opportunity we can’t afford to miss.” “We’re missing the boat.” “We’re becoming redundant.” “We’re not in control.” Business owners and executives are asking themselves how to deal with this AI tsunami. If this is the real thing, what should we do? And if it isn’t, can we just carry on as before?
I have been looking around inside organisations for thirty years. I talk with managers, executives and board members, and also with high-tech experts. My conclusion: the truth lies somewhere in the middle. You can’t leave everything as it was, and you can’t leave everything to AI either. Sometimes you have to turn things around to see what is really going on.
The buzzword of the moment is: let AI do the work and keep people in the loop. But can we afford that?
With powerful, predictable machines it works fine. A crane or a transport vehicle does the work, and people monitor its behaviour and output. You know how the machine works, and you notice when something is off.
But when you can no longer understand a machine, its outcomes become unpredictable. And when that machine evolves organically, without you knowing that, how or why the process is changing, its behaviour becomes impossible to pin down. Suddenly the outcome is different, and you can’t work out why. That happens with many AI applications. The human as an inspector at the end of the line then becomes a formality, or a brake that eats up all the gains.
My field research led to the concept of Human-Centric Process Management with AI “in the loop”. Not people in the loop of AI, but AI in the loop of people. The business process, designed by people, is the starting point. Within it, AI is deliberately used where it adds value.
The core idea: as a team, be the architect of your business processes and of the connections between them. As a team, because operational staff, managers and specialists together know how the work really flows. And the connections, because handovers between processes are exactly where things most often go wrong. An AI application that optimises one process can quietly disrupt the next.
AI operates within three frameworks that people set and safeguard:
These frameworks are not an add-on for the compliance department. They are how the organisation steers. As long as people set and safeguard them, you stay in charge of AI.
Within these frameworks, AI gets plenty of room:
Because AI works within a design that people understand, outcomes remain explainable. If something changes, you know where to look.
From executives it requires direction: make the purpose and goals explicit. From managers it requires process thinking across departmental boundaries. From professionals it requires ownership: your knowledge of the work is exactly what AI lacks. And from IT and AI specialists it requires putting their skills at the service of the process design.
Choose one value stream that matters, including its connections. Make the purpose, goals and stakeholders explicit. Together with the people who do the work, map out how the process really runs. Design the desired process, and only then decide where AI can help. Agree on who the owner is and how you will keep monitoring AI. Learn, and scale up.
How do you, as an organisation, stay in charge of AI? Not by slowing AI down, and not by leaving everything to AI. But by keeping firm ownership of your processes. Opportunism and fear are poor advisers. Design capability and ownership are better ones.
Build human-centric business processes with AI in the loop!
Discuss them with an expert or book an appointment »