AI Won't Fix Broken Processes: A Key Takeaway from Money20/20
One of the most thought-provoking AI sessions at this year's Money20/20 Europe was "From Hype to Habit: Eight Lessons for Leading in the Age of AI", presented by Marguerite Bérard, CEO of ABN AMRO Bank NV. Drawing on the experience of leading a 200-year-old European bank through the AI era, she shared eight practical lessons on how organizations can move from AI experimentation to responsible adoption at scale.
Among all the insights discussed, one message stood out: AI cannot compensate for weak foundations. If systems are fragmented, processes are inefficient, or data is unreliable, AI will not solve those problems—it will simply amplify them. This article explores that idea and why it may be the most important lesson for organizations preparing for the AI age.
AI Won't Fix Broken Processes
Artificial Intelligence dominated the conversations at Money20/20. New tools, new capabilities, and new use cases were everywhere.
Yet one of the most valuable insights from the stage was surprisingly simple:
"If you do not have robust IT, good processes, and relevant data, AI won't be able to do anything for you."
In a world increasingly focused on AI adoption, this statement serves as an important reminder: technology alone is not enough.
The AI Paradox
As highlighted during the presentation, every major technological revolution follows a similar pattern.
In the short term, expectations are often enormous. Investment pours into new technologies. Headlines predict dramatic transformation.
At the same time, the actual impact on productivity and business performance often appears modest.
This is not because the technology lacks potential.
It is because realizing that potential takes time. Organizations need to adapt, learn, and integrate new technologies into the way they operate before meaningful value can be created.
AI is no different.
AI Adoption Is an Organizational Challenge
One of the strongest themes throughout the presentation was that AI success depends on how effectively it spreads throughout an organization.
This requires leadership involvement, employee engagement, education, governance, and experimentation.
Organizations cannot simply declare AI a priority and expect transformation to happen.
Leaders must actively participate, learn, and demonstrate commitment.
Employees must understand not only how to use AI, but also how it affects their work.
Risk, compliance, and audit functions must be included in the journey.
AI adoption is not a technology project.
It is an organizational change initiative.
Human + AI Is Better Than Human or AI Alone
A particularly interesting example shared during the presentation involved university students working with AI on a forecasting assignment.
Three groups emerged.
The first group simply accepted AI-generated answers.
The second group used AI only to confirm what they already believed.
The third group actively challenged, refined, and expanded their thinking through interaction with AI.
The result was striking.
The strongest outcomes came from those who used AI as a thinking partner rather than a replacement for thinking.
This reflects a broader lesson for organizations.
The future does not belong to teams that outsource their judgment to AI.
It belongs to teams that learn how to think with AI.
The winning formula is not human alone.
It is not AI alone.
It is human plus AI.
Why Foundations Matter
Perhaps the most practical lesson from the presentation was the emphasis on discipline and systematic execution.
AI cannot compensate for weak foundations.
As the speaker explained:
"Try to plug good AI on a bad process, you still have a bad process."
Organizations often focus on AI use cases before addressing the fundamentals.
However, sustainable AI adoption requires:
Robust IT foundations
Well-designed business processes
Relevant and reliable data
Scalable operating models
Without these elements, AI initiatives risk remaining isolated experiments rather than drivers of meaningful business value.
Scaling AI Requires More Than Experimentation
Another important distinction made during the presentation was the difference between isolated success stories and enterprise-wide impact.
Organizations should not focus solely on creating individual AI use cases.
The goal should be to create repeatable patterns that can be deployed across multiple functions and processes.
Real value emerges when AI becomes part of the organization's operating model, not when it remains a collection of disconnected pilot projects.
Looking Ahead
The presentation concluded with an optimistic perspective.
AI will undoubtedly challenge existing business models, create new risks, and force organizations to rethink how they operate.
But if implemented thoughtfully, it also has the potential to help organizations become more effective, more adaptive, and ultimately better institutions.
The message from Money20/20 was clear:
AI is not a shortcut around organizational challenges.
Organizations that want to benefit from the AI revolution must first ensure they have the right foundations in place.
Because successful AI adoption starts long before the first AI tool is deployed.
Photo: Money 20/20