AI won't save your supply chain. Your data will.
Supply chains have never been more complex, volatile and fragile. The challenge isn't disruption itself, because unpredictability has always existed. It's the speed, scale and frequency of today's shocks.
With global unrest closing key trade routes, I keep coming back to one question when I talk to supply chain teams: how confident are they that they really understand the risks sitting in their operation, and how to mitigate them?
Over the years I've had the privilege of speaking to a lot of teams who are getting this right. Here's what I've noticed they have in common.
Your best insights are already there. They're just trapped
The best performing teams I meet have built something that's genuinely hard to replicate: institutional knowledge of how their operation actually behaves. That accumulated understanding, most of it earned through the lived experience of operators, is what lets them know which suppliers reliably meet lead times, which lanes quietly underperform, and where costs leak each quarter.
The trouble is that this knowledge is trapped. It's scattered across spreadsheets, inboxes, freight portals, carrier scorecards and people's heads. None of these systems were built to talk to each other, and depending on where you look, they'll often give you a different answer. The insight exists. It's just hard to get to.
That's why I think there's a real case for consolidating and standardising this data through integrations and APIs. Connecting siloed data isn't just tidier. It gives you the context that matters when you're trying to de-risk. Instead of only seeing that a shipment is late, a team can see who owns it and what that delay is likely to cost.
AI is only as good as the context behind it
You might expect me to say AI is the obvious next step. And it matters. But its value rests entirely on the data foundation and operational context underneath it. Once a team's data is connected, AI can start reasoning over patterns in their specific context rather than generic ones. Suddenly the whole team can see recommendations that used to take years of experience to form.
The compounding asset here is the context layer. Every quarter of clean data, every performance review that captures what actually happened rather than what was promised, makes the intelligence sharper. The longer that context builds, the harder it becomes for anyone to replicate from scratch.
So when it comes to de-risking, I'd argue it's less about which AI models you use and more about how you develop the operational context that underpins them.
Better data changes the question entirely
Once data and context are consolidated, the decision-making model shifts. Instead of spending their time reconciling numbers, teams can start asking better questions.
In the past, a team might only have wanted to know where a shipment or a unit of stock was at a given moment. That's useful. But in a volatile landscape, asking why a lane keeps falling short, and what specifically that is costing, tends to be far more effective at reducing risk. It's a fundamentally different conversation.
With that kind of insight, leaders can spread risk and reduce their dependency on any single route or carrier. They can see which lanes are performing better, and pivot to back-up options quickly when something goes wrong.
The divide is coming
AI gets cited as the answer to de-risking supply chains. I think that's only half right. What really matters is the data connectivity and infrastructure underneath it. Connect your data, build a compounding operational context for AI to run on, and you can start asking granular questions and making decisions that genuinely mitigate risk and cost.
My honest guess is that over the next few years we'll see a divide open up. On one side, teams using AI on clean, proprietary operational data to make decisions that couldn't previously be predicted. On the other, teams relying on the same data everyone else has, and quietly wondering why the edge they hoped for never arrived.
Supply chains are going to stay uncertain. Which is exactly why there's never been a better time to build this foundation and put your competitive advantage to work.