AI alone won't transform your supply chain. The data beneath it will.

The companies that will win the artificial intelligence (AI) race in supply chain aren't the ones with the most advanced AI models. They're the ones with the deepest operational data, the strongest domain expertise and the most connected ecosystems. 

There is a seductive simplicity to the way most enterprises talk about artificial intelligence today. Deploy a large language model. Automate a few workflows. Announce the transformation. Move on. 

Supply chain leaders know better. They have been through enough technology cycles to tell a genuine shift from an over-hyped or well-marketed one. EDI, cloud ERP, digital twins, blockchain - each wave brought its share of believers and sceptics. The early evidence from those building AI seriously into the fabric of global logistics suggests the stakes this time are different. The gap between companies that get it right and those that don't will be vast, structural, and very hard to close. 

The reason comes down to three things: operational data, network strength and domain expertise.  

The interface trap 

When Zubin Appoo, CEO of WiseTech Global, describes how the company thinks about AI, he doesn't start with the technology. He starts with the problem to be solved, and specifically with the customer on the other end of it 

Appoo explains, "When a freight forwarder is managing hundreds of shipments through a geopolitical crisis, or a customs broker is working through complex classification decisions under mounting regulatory pressure, the last thing they need is to go hunting for information. AI built into the core of their platform means the right information finds them, so they can act rather than search". 

That distinction is more important than it first appears. Much of supply chain enterprise AI adoption today is still happening at the interface level. A chatbot interface sitting over a legacy system. An automation tool scraping screens and filling forms. A copilot bolted onto a platform that was never designed for it. 

These tools can be useful. They can reduce friction, speed up repetitive tasks and make systems easier to use, but they do not fundamentally change how the operation works. 

Surface-level AI takes the existing workflow as given and makes it faster. 

Transformative AI asks whether the workflow should exist in its current form at all. It looks at the underlying process, the data that drives it, the decisions it requires and the execution steps that follow. Then it helps redesign the operation around what is now possible.  

That is the difference between AI as a feature and AI as infrastructure. 

Why operational data is becoming a competitive advantage 

One of the biggest misconceptions in enterprise AI is that the model itself is the differentiator. In reality, it isn’t. 

Large language models (LLM) from OpenAI, Anthropic, Google, and others are increasingly commoditized. The capability gap between them narrows with every release cycle. What cannot be commoditized is the proprietary data that gives those models something real to work with. 

"Data becomes king," Appoo explains. "The only way we can make intelligent decisions here is by applying that LLM layer to that data." 

For WiseTech, those data sets have been accumulating for 32 years. It spans customs filings, freight movements, carrier connections, compliance records, and operational workflows across countries that represent 80% of global trade flows. When a large language model is applied to that kind of operational ecosystem, the outputs become fundamentally different from what generic systems can produce. 

Consider a major trade disruption affecting a critical global shipping corridor. A generic AI tool may be able to summarize news reports, identify affected regions or surface broad guidance on alternate routes. That can be helpful, but it is not enough for operational decision-making. 

A logistics platform built on shipment history, booking data, carrier behaviour, routing patterns, compliance rules and execution workflows can go further. It can identify which shipments are exposed, suggest realistic alternate routes, estimate transit time impacts, assess cost implications and help teams decide what to do next. 

That is not simply information retrieval. It is operational intelligence. 

This is where proprietary data becomes a strategic advantage. AI models can be licensed, accessed and integrated. However, deep operational data cannot be replicated overnight. Neither can the domain knowledge required to interpret that data correctly in the context of global trade. 

Building intelligence through networks and shared data 

Data at scale matters, but the supply chain is not just a data problem. It is a network problem. 

Let’s take an example of a manufacturing supply chain. A production run depends on components arriving on time. Those components move through a chain of interdependent handoffs between suppliers, freight forwarders, customs brokers, shipping lines, terminals, inland carriers and port authorities before they reach the factory floor. A delay at any one point is not an isolated exception. It is a production halt, with machinery idle and costs compounding by the hour. 

In a fragmented environment, each participant may see only their part of the problem. In a connected environment, those signals can be understood together. A freight forwarder operating within a connected platform does not just see the customs delay. They can understand what it means for downstream production, customer commitments and alternate execution options. 

This is what network density makes possible. Every additional node connected to a platform, whether a carrier, customs authority, warehouse, terminal, forwarder or manufacturing facility, adds context to every other node. The system becomes meaningfully smarter with each integration, not simply broader. The value of any platform is a direct function of the number of nodes it connects to. 

WiseTech’s recent acquisition of e2open extends its reach beyond logistics execution into a broader multi-enterprise supply chain ecosystem, connecting 500,000 enterprises across manufacturing, logistics, channel and distribution networks. That is an important step because AI becomes more powerful when it can reason across a wider set of trading relationships, workflows and operational signals. 

But integration depth is only half the equation. AI is only as intelligent as the data flowing through it, and too much of that data is still trapped in fragmented systems, re-entered at every handoff and reconciled manually at every boundary.  

Harnessing the true power of AI requires trusted data flow across the ecosystem, shared through open, secure, standardised channels. That starts with a simple principle: enter data once, enter it accurately, allow it to flow end-to-end across the shipment lifecycle, and use it to drive better decisions. 

For this to work, the industry needs deeper collaboration between governments, carriers, forwarders, customs brokers, logistics providers, industry groups and technology partners. The goal is not just better data exchange, but fewer duplicated processes, fewer manual handoffs and a more connected foundation for how global trade operates.  

The human expertise behind effective AI 

AI also needs something that is easy to underestimate: domain expertise and human judgement. 

Global trade is not a generic workflow. It is a highly regulated, deeply localized, and operationally complex ecosystem. Customs processes vary by country. Carrier practices vary by region. Regulatory requirements change frequently. Documentation rules, tariff logic, routing constraints and execution realities are shaped by years of operational context. 

That context cannot be inferred from an AI model alone. 

Appoo frames WiseTech’s acquisition strategy through this lens. The company has acquired nearly 60 businesses over the last decade, not simply for their technology, but for the domain expertise and operational knowledge behind them. "We're acquiring people who know what Belgium customs looks like because that's what they've lived and breathed for the last 15 years." 

That expertise matters because global trade is full of decisions where the difference between “technically possible” and “operationally right” is significant. AI can surface patterns, risks and recommendations, but domain expertise helps determine what those signals actually mean, which actions are compliant, and which decisions can be executed in the real world. 

Turning intelligence into action 

Data gives AI the raw material. Network density gives it context. Domain expertise helps it understand what those signals mean. 

But in supply chain, intelligence only matters when it leads to action. 

The real test of AI-driven supply chain platforms is whether they can help teams decide what to do next and then execute that decision across systems, partners, and workflows. 

Disruption is no longer an occasional exception in global trade. It is part of the operating environment. The question is not whether supply chains will face the next shock, but whether organizations will be ready to respond when it happens. 

That readiness depends on three things working together. 

The first is richer inputs. No single data feed is sufficient anymore. Meaningful risk intelligence requires combining real-time news, sanctions data, weather signals, congestion indicators, satellite and vessel movement analytics, carrier operational data, tariff information, and booking records. Individually, those signals are noisy. Together, they reveal patterns that can identify sovereign risk, route infeasibility, and supply chain exposure long before it makes the headlines. 

The second is integrated risk modelling. Collecting data is not enough on its own. Organizations need systems that convert that data into clear, scenario-based decisions. Not simply an alert that says a route is at risk, but a clear operational response. If this corridor becomes unavailable, reroute here. If this port is congested, consider this alternate gateway. If this regulation changes, trigger this compliance workflow. The response plan should exist before the crisis begins. 

The third is execution capability. Insight without execution is just a well-designed dashboard. A risk alert is of limited value if the organization cannot act on it rapidly. Systems need to trigger communications, amend bookings, switch ports, arrange alternate transport, and coordinate stakeholders in near real time. That requires automation, governance, and pre-approved operational playbooks rather than improvisation under pressure. 

This is where AI becomes most powerful. It can ingest millions of data points continuously, prioritize what matters, and help teams make faster, more confident decisions at a scale no human team could match, freeing people to focus on what they do best: customer relationships, judgment, negotiation, and complex problem solving. 

What this means for every supply chain leader 

The window for building this kind of data and network advantage is not infinite. The organizations investing now in embedding AI into their operational architecture, rather than bolting it on top, are compounding a lead that will be very difficult to close in three to five years. 

For leaders evaluating their own AI strategy, the right questions are not about which model to use, or which vendor has the best interface. The right questions are structural. 

Is our AI working with proprietary operational data, or generic inputs? Can it access business logic and underlying architecture, or only what a human user can see on a screen? Are we building genuine network depth through more integrations, more data sources, and deeper collaboration across the ecosystem?  And are we using AI to redesign how we operate, or simply to run the same operations slightly faster? 

"AI means we can solve the problems that need solving more deeply, faster, and possibly solve problems that we couldn't solve before," Appoo says. "I would never talk about AI as a new offering. For three decades we solved efficiency and solved risk. AI means we can do that far greater than we ever could before." 

That is the thinking of an organization that has moved past the hype. AI is not a destination. It is an accelerant, and what it accelerates depends entirely on the foundation beneath it. 

The supply chain is the circulatory system of the global economy. The companies that treat their intelligence layer with the same seriousness they bring to their physical infrastructure will define the next era of trade. The question is whether your organisation is building that layer or is still shopping for a plug-in.