Sleepwalking into AI? Why retail innovation still depends on the grey eminence

I’ve previously written about the risk of sleepwalking into the cloud—when companies adopt new technology simply because others are doing so, without a clear strategy for how it will create real value.

After attending this year’s D Congress in Gothenburg, Sweden, I found myself asking a similar question: Are we now at risk of sleepwalking into artificial intelligence (AI)?

AI and AI agents were everywhere. But unlike previous years, which were dominated by theoretical presentations and visionary concepts, this time it felt tangible. Companies are no longer just talking about AI; they are actively experimenting, building, and deploying solutions. The technology is becoming easier to use, and new applications can be developed faster than ever.

The rapid adoption of AI is exciting, but also raises an important question: What makes it work beneath the surface?

The invisible foundation of retail

At conferences like D Congress, it’s easy to focus on visible innovation—AI powered services, new customer experiences, and rapid application development. Yet behind every successful innovation lies the same fundamental truth: Without structured data, reliable systems, and a robust architecture, even the most advanced front end solutions fail to deliver value.

That’s why I often describe companies like Infor™ as the grey eminence of retail—the influential force behind the scenes. We may not always be the most visible part of the technology stack, but the systems that manage supply chains, inventory, logistics, and finance form the backbone of everything else. AI agents and digital services are only as effective as the data and processes behind them.

AI as a layer, not a replacement

One example from the conference illustrated this clearly. A mid sized e commerce company described how they test new functionality internally using modern development tools and AI assisted coding. If a solution proves effective, it is deployed. If not, the company turns to external platforms.

This approach enables rapid innovation at the edge, while core systems continue to provide structure, stability, and control. In many cases, this may represent AI’s most effective role in the near term: Extending enterprise systems rather than replacing them.

Don’t sleepwalk into the next wave

This is where the lesson of sleepwalking into technology becomes relevant again. It’s easy to be captivated by the speed and simplicity of modern AI tools—and that excitement is justified. Many organizations can now build in hours what once took months.

But speed is not strategy.

The retailers that succeed with AI won’t be the ones that experiment the most. They will be the ones that balance experimentation with governance, high quality data, and a clear architectural direction. Without that foundation, AI risks becoming just another layer added to already fragmented systems.

Retail is evolving at an unprecedented pace, and AI will undoubtedly play a central role in that transformation. Yet the principle remains unchanged: Innovation at the surface creates value only when the structure underneath is strong.

In other words, the future of retail innovation still depends on the grey eminence of robust, well-governed back-end systems.