IQAX interview puts data interoperability at the center of logistics AI

The document layer beneath logistics AI

In an interview with TechNode, IQAX CEO George Guo described a strategy that connects electronic bills of lading with cargo monitoring, finance and operational decisions. The company says it has processed more than one million electronic bill-of-lading transactions, or eBL transactions, using infrastructure built on the GSBN blockchain network.

IQAX also describes an IoT footprint of about 92,000 connected devices spanning more than 200 regions and 11,000 city pairs. Its proposed AI applications include arrival predictions, cold-chain monitoring and identifying disruptions early enough for operators to change plans. These scale figures come from the company, not an independent audit.

Different technologies perform different jobs

The useful distinction is between a record, an observation and a prediction. A shipping document records rights and transaction details. A sensor observes a physical condition such as temperature. A model estimates what may happen next. Combining them can help an operator understand which cargo is affected by a delay, but a prediction does not replace the underlying document or inspection.

Interoperability is therefore central to the proposal. If a carrier, bank and customer describe the same shipment differently, an AI system can produce a fluent answer without resolving that mismatch. Consistent identifiers, permissions and an agreed history of changes are prerequisites for actions that cross company boundaries.

The data behind the digital twin

IQAX's own technology page describes a digital-twin platform combining more than 40TB of schedules, sensor signals and operating data. Its data-services page identifies more than 20,000 vessels, 1.6 million port-pair schedules and over 1,000 ports in its stated coverage. These are company descriptions of data scope, not a claim that every shipment has equally complete information.

The same product documentation distinguishes carrier schedules from other observations. A published arrival time is a plan; a vessel-position signal provides evidence about actual movement. Combining them can support a revised estimate and flag a likely disruption. The important step is turning conflicting or incomplete inputs into a forecast that an operator can check, rather than simply accumulating more records.

IQAX's May 27 announcement dates its one-million-transaction milestone earlier than the September interview. That matters because the interview is a discussion of an established rollout, not the date on which the company first crossed that threshold.

How a customer gets shipment data

IQAX publishes an English shipment-tracking API specification that makes the integration more tangible. Customers first obtain a subscription key and register shipments. Registration does not immediately guarantee a usable response: the documentation provides a status check, and a shipment must reach its Ready to Track state before tracking results are available.

The service offers both requests initiated by the customer and webhook notifications sent when updates occur. For notifications, the customer must operate a reachable server and arrange the callback address before enabling the service. A forwarding company could therefore connect shipment events to its existing operations software instead of asking staff to inspect a separate dashboard.

These steps establish an available integration interface, not a guarantee about predictive accuracy. They also identify work that remains with the customer: registering the right consignments, receiving updates reliably and matching each event to its own shipment records. That practical dependency helps explain why the interview gives data interoperability such a prominent role.

Read the performance claims as case studies

IQAX cites a Peru-to-China grape shipment that took 23 days, ten days less than a traditional route, with a reported 20% reduction in transport costs. It also says AI can reduce some pre-trip inspection work by up to 70%. The interview does not provide a controlled comparison separating the effects of routing, planning and AI.

Those examples make the intended outcome concrete, but they should not be treated as guaranteed savings for another lane or cargo type. For an operator, the meaningful test is whether earlier warnings change a decision and reduce a measurable loss. Arrival-prediction accuracy, false alerts and the cost of acting on them would help distinguish a useful logistics tool from a dashboard that merely displays more information.