Freight Audit Is Broken: Real-Time AI Fixes It

Freight audit is broken when you’re finding billing mistakes a month later. This is how real-time AI freight audit catches shipment cost errors as they happen. Shipium’s Jason Murray and Penny Allen break down Always on Audit, why monthly invoice reviews miss the real problem, and how continuous auditing ties the shipment record to the invoice, disputes and credits in real time. Strong discussion on digital twins, messy carrier data, AI agents and the gap between what a shipment should cost and what shows up on the bill. #FreightAudit #SupplyChainTech #ParcelShipping

Shipium has launched Always On Audit, an AI-driven freight audit product designed to replace the traditional monthly batch review of carrier invoices with continuous, real-time auditing at the shipment level. The product targets a long-standing pain point for shippers: by the time a billing discrepancy surfaces in a monthly audit cycle, credits can be weeks old and root causes are nearly impossible to trace. The company says the new tool is already catching errors — including one beta customer being charged base rates instead of contracted rates, costing “many thousands of dollars” before Shipium intervened.

Shipium CEO and co-founder Jason Murray said the company stayed out of the freight audit space until 2022, when advances in generative AI and agent-based systems made a fully automated, real-time approach viable. The platform deploys AI agents running around the clock to compare every invoice and receipt against a versioned “digital twin” — a gold-standard record of what each shipment was supposed to cost, built from the rating and routing decisions Shipium already makes upstream.

“The best defect being one that never gets into the product — and the same thing applies here,” said Penny Allen, SVP of AI Product and Engineering at Shipium, describing the architectural goal. “The best-case scenario is a bad charge never makes it onto the invoice.”

“Monthly audit and continuous audit are not the same job at different speeds. They audit different objects. Monthly audit is a batch job over one artifact, the invoice. It’s archaeology. Continuous audit’s object is the shipment itself.” — Penny Allen, SVP of AI Product and Engineering, Shipium

Allen said the system’s forensic capability goes well beyond flagging a dollar discrepancy. When an accessorial charge slips through, the platform can identify not just the date and time, but the lane, the dimensioner, the station, the operator, and the data stream that introduced the error. In one early beta case, the system detected that a carrier had begun applying a surcharge two days before the contractually specified start date and corrected it immediately. The platform also monitors for slow-moving drift — cost creep on a lane, a quietly shifting carrier mix, or a rate card that has diverged from reality — before those anomalies grow large enough to trigger a human review threshold.

Murray, who spent approximately 19 years at Amazon before founding Shipium in 2019, said the core problem he observed there was structural: vendors and carriers would receive chargebacks for errors but often lacked the tools to trace and fix the upstream cause. Always On Audit is designed to close that loop by handing findings directly to an agent that attempts to correct the problem automatically and escalates to a human only when it cannot. Each human resolution, Allen said, then trains the system to handle the same exception autonomously in the future.

On the broader Shipium platform, the company says it typically delivers a 12% reduction in parcel transportation costs for customers. Murray noted that around 2023, Shipium crossed roughly $100 million in coverage volume in the U.S., the scale at which its predictive models materially outperformed standard carrier-provided estimates. Always On Audit is currently rolling out to existing parcel customers, with plans to expand to additional modes and international markets. The company said implementation begins generating findings immediately, as the digital twin starts populating from the first moment of onboarding.

• Shipium’s Always On Audit uses AI agents running 24/7 to audit shipments in real time, replacing traditional monthly invoice batch reviews that delay error detection by weeks.
• In early beta tests, the system caught a carrier billing base rates instead of contracted rates — costing a customer many thousands of dollars — and flagged a surcharge applied two days before its contractual start date.
• Shipium reports a typical 12% reduction in parcel transportation costs across its platform and reached roughly $100 million in U.S. coverage volume around 2023, the scale it needed for its predictive models to outperform carrier-standard estimates.

This Summary is generated thanks to a transcription of the interview, for the full interview please enjoy the video above.