AI Spending Still Leaves Tariff Management Exposed

Supply chain operators are pouring money into artificial intelligence, yet tariff management remains a stubborn blind spot — a gap the industry has not closed despite the investment wave.

That is the blunt assessment from Supply Chain Brain, which argues that AI budgets across logistics are rising while the specific problem of managing tariffs — classification, duty exposure and compliance — keeps slipping through.

It sounds counterintuitive. AI should be exactly the tool for tariff work. Classification codes, rules of origin and duty rates are data-heavy, rules-heavy and repetitive. Machines read rules well. So why hasn't the money moved the needle?

Where the money goes — and where it doesn't

The short answer: AI investment has flowed to the visible, high-volume parts of the stack. Forecasting. Visibility. Route optimization. Chatbots.

Tariff management sits deeper in the workflow. It touches product data at the SKU level, supplier declarations and customs documentation — the messy, fragmented layer where master data is often wrong or missing.

No algorithm fixes bad input. If the product master file lacks material composition or country-of-origin detail, the smartest classification engine still produces an answer built on sand.

What does that mean for shippers?

For shippers and forwarders, the practical consequence is simple. Duty exposure remains a manual, error-prone process even at companies that describe themselves as AI-forward.

That carries real cost in several ways:

• Misclassification risk, with penalties and retroactive duty bills
• Slow response when tariff schedules change mid-cycle
• Overpayment of duties that no one audits
• Compliance exposure that grows with each new trade lane

The gap matters most for importers running multi-origin sourcing, where the same SKU can carry different duty treatment depending on where it was made and how it was declared.

A data problem wearing an AI costume

The core issue, as the analysis frames it, is that tariff management is not really an AI problem. It is a data governance problem.

AI can accelerate classification once the underlying product data is clean, complete and maintained. Before that point, automation simply speeds up the propagation of errors.

That distinction is where investment strategies go wrong. Companies buy tools. They don't fund the unglamorous work of fixing master data, mapping supplier declarations and keeping origin documentation current.

What comes next

Expect the focus to shift. As tariff volatility persists and customs authorities tighten enforcement, operators will face pressure to close the gap from the data side first — cleaning product records before layering intelligence on top.

The lesson for supply chain leaders: AI spend without data foundations won't close the tariff-management gap, and 2025's budget cycles will likely have to account for that.