AI: What It Won’t Do (and Why That’s Actually the Exciting Part)

What a year this is shaping up to be for AI — it’s showing up in almost every conversation I’m having across industries.

To fully geek out for a moment, I was listening to a podcast recently that had nothing to do with freight, and yet AI naturally came up in the discussion. I also attended a networking dinner this week where I spoke with people from half a dozen completely different industries, and once again — AI was everywhere in the conversation. There was talk about where people use it, how they use it, how they recognise it, and just as importantly, where it doesn’t work.

That last part is what stood out.

No matter the industry, and no matter how different the use cases are, one idea keeps resurfacing:

Garbage in, garbage out.

Not as a limitation — but as a reminder that the real power of AI is directly tied to the quality of the workflows and data it’s built on.

This principle has been around long before artificial intelligence, but right now it’s becoming more relevant than ever.

AI is already starting to reshape what we can do in logistics — faster document processing, smarter reporting, automated communication, and improved visibility across operations. And this is only the beginning. As these capabilities continue to make their way into the logistics platforms we use every day, the opportunity is significant.

But here’s where it gets interesting.

AI doesn’t operate in isolation. It reflects the environment it’s given.

If shipment milestones are inconsistent, if reference data is incomplete, or if teams apply processes differently across branches, AI doesn’t fix that — it amplifies it. What used to be small inefficiencies become visible at scale. What used to be manual workarounds become embedded outputs.

This is especially relevant within CargoWise environments, where the platform is already deeply workflow-driven. Every status update, document upload, exception note, and task completion contributes to the structure AI will eventually rely on.

And this is where the real opportunity sits.

In practice, workflow capabilities within CargoWise are not always fully leveraged or applied consistently across teams and branches. Existing workflows may not have been reviewed recently to ensure they still reflect current ways of working, and in some cases workflows may not be used at all — creating a clear gap between system capability and real-world execution.

That gap is exactly where AI will have the most interesting impact, and where the biggest gains are likely to be found.

As AI becomes more deeply embedded into logistics software, workflow design is shifting from an operational detail to a competitive advantage.

The organisations that benefit most from AI won’t necessarily be the ones that adopt it first.

They’ll be the ones who already understand their workflows — who have invested in consistency, clear ownership, and accurate data capture. In that environment, AI doesn’t just automate tasks — it accelerates performance.

It removes friction. It reduces administrative load. It gives teams back time to focus on higher-value decisions and problem-solving.

AI doesn’t replace discipline — it rewards it.

So before asking, “What can AI do for us?” the more powerful question might be:

What kind of system are we building for AI to work within?

Because in this next phase, the real advantage won’t come from who uses AI — it will come from those who give it what it needs to perform at its best.