Descartes Systems Group claims its new AI agent can cut global trade data research time by as much as 90% — and it is betting that shippers, forwarders and sourcing teams will pay for the shortcut.

The company launched the Descartes Datamyne AI Agent on Sept. 24, embedding a natural-language research tool directly into its global trade intelligence platform. Users type questions about suppliers, buyers, commodities, shipment activity and trade relationships. The agent returns contextual answers, visualizations and — critically — the underlying shipment records that support each finding.

That verification layer is the commercial pitch. Ken Wood, executive vice president of product management at Descartes, framed the tool as AI married to domain-specific data rather than a general-purpose chatbot.

"The Agent analyzes shipment-level data and presents findings alongside the records and visual evidence supporting them," Wood said in a statement. "Embedding the Agent directly into Descartes Datamyne gives customers a more transparent and intuitive way to accelerate decision-making and respond more effectively to changing global trade conditions."

What sits underneath

The dataset is the moat. Datamyne covers 230 markets and adds more than 500 million shipment records annually, according to Descartes. The database lets users identify suppliers, monitor competitors, evaluate trade risks and track shifts in product flows and market demand.

Until now, interrogating that data required building database filters and learning specific query structures. The agent removes that barrier. Users can ask a question in plain language, then keep refining it and drill into related companies, commodities, shipments and trade relationships without leaving the platform. The tool can also generate summaries that flag trends, anomalies and potential sourcing opportunities across large volumes of trade records.

The 90% figure carries caveats. Descartes did not explain how it calculated the time-reduction estimate and qualified the claim by noting that results depend on user experience and query complexity. Buyers evaluating the tool should treat the headline number as an upper bound, not a baseline.

Third AI release in three months

The Datamyne launch extends a rapid AI buildout across Descartes' logistics software. In August, the company introduced AI-powered image document management to help customs brokers, freight forwarders and logistics service providers extract information from trade documents and prepare data for customs and transportation processes. Earlier that month, it released an AI-enabled Free Trade Intelligence solution designed to help manufacturers, importers and exporters identify opportunities to reduce trade costs.

The pattern is deliberate and mirrors the wider freight-tech market: vendors are embedding AI into existing software and proprietary datasets rather than selling standalone chatbots. Descartes has effectively bolted a conversational interface onto Datamyne's trade data — a lower-risk path to AI monetization than building a new product from scratch, because the underlying records and the customer base already exist.

For trade compliance and sourcing teams, the practical consequence is a shift in how shipment data gets used. Analysts who once depended on structured queries can now run exploratory investigations — a competitor's supply chain, a commodity's flow through a specific lane, a buyer's import activity — in a fraction of the time. That compresses the research cycle for supplier vetting, trade-risk screening and market-demand tracking.

The investment backdrop

The launch lands amid a decade-long surge in transportation technology spending. Descartes' annual transportation management benchmark survey, released shortly before the Datamyne announcement, found that transportation technology investment has risen nearly 50% over the past ten years. The company said the next phase of adoption will hinge partly on companies' ability to scale automation and artificial intelligence.

That framing positions the AI agent as more than a convenience feature. If trade-intelligence research collapses from hours to minutes, the competitive value of shipment-data access rises correspondingly — and vendors that combine proprietary datasets with usable AI interfaces stand to capture a disproportionate share of the sector's growing software budgets.

Descartes has not disclosed pricing or adoption targets for the agent, and the effectiveness of the 90% claim will only become clear as users test it against real queries on live lanes. What is clear is the trajectory: the company has shipped three AI products in three months, and each one points toward AI becoming the default interface for freight and trade data rather than an add-on.