For supply chain leaders, the difference between a healthy freight spend and one riddled with hidden leaks often comes down to the ability to detect small, seemingly innocuous anomalies before they compound into six- or seven-figure losses. Intelligent Audit's DeepDetectAI, built by Chief Product Officer Brian Pollack, is a proprietary machine learning system designed to do exactly that — earning the company an honoree spot in FreightWaves' AI Excellence in Supply Chain Awards in the AI Solution Provider category.
According to FreightWaves, the awards recognize real deployments and measurable outcomes, judging entries on the strength of the AI application, integration into workflows, and tangible results. DeepDetectAI was singled out among these criteria.
The system starts by ingesting a shipper's full transportation data set and using machine learning to establish a baseline of normal shipping activity for that specific business — down to the account, service, and geography level, according to Intelligent Audit. From there, it continuously monitors new activity for cost variations, unusual service usage, duplicate or fraudulent charges, and other deviations from that baseline.
What separates DeepDetectAI from a standard alerting tool, according to Intelligent Audit, is the explainability layer: every anomaly comes with data showing what happened, where it occurred, and why, backed by support from an Intelligent Audit analyst. This combination, the company says, moves logistics, finance, and operations teams from reactively digging through data to proactively resolving exceptions as they surface.
Case Studies: Small Signals, Compounding Consequences
Intelligent Audit's body of DeepDetectAI case studies reveals how small, easy-to-dismiss deviations can compound into large problems. A pattern observed across multiple accounts.
| Client Type | Anomaly Detected | Outcome |
|---|---|---|
| Global eyewear giant | $10,000 spike in unused return service | Revealed organized fraud ring using manipulated barcodes and compromised UPS accounts; over $1 million in fraudulent activity identified; triggered FBI investigation |
| National specialty retailer | Service-selection error: FedEx Home Delivery booked instead of Ground across multiple accounts | Would have resulted in millions in avoidable spend if not caught |
| Global multi-brand manufacturer | New "Additional Classification Fee" on non-authorized brokerage account | Flagged before it compounded weekly into over $200,000 of unplanned spend |
| High-end fashion retailer | Three separate signals: late-payment fee spike, first-time use of premium expedited service, incorrect international freight charges | Multiple small anomalies stacking up inside a narrow window |
According to FreightWaves, in the global eyewear giant case, the $10,000 spike in a return service the company had never used before turned out to be the first signal of an organized fraud ring buying glasses, manipulating return barcodes, and compromising UPS accounts to reroute product from Mexico into the U.S. By the time the full scheme was uncovered, more than $1 million in fraudulent activity had been identified, prompting an FBI investigation.
For the national specialty retailer, teams were unknowingly booking FedEx Home Delivery instead of Ground across multiple accounts — a mistake that would have cost millions. For the global multi-brand manufacturer, a single new "Additional Classification Fee" billed to a non-authorized brokerage account was flagged before it could grow weekly into over $200,000.
What This Means for Your Procurement Team
The practical implication for chief supply chain officers and procurement leaders is clear: freight and parcel invoices are some of the densest, highest-volume data a shipper touches, according to Intelligent Audit. A single enterprise account can generate thousands of line items a week across carriers, services, accessorials, and billing cycles. That volume, the company notes, is exactly where costly mistakes and deliberate fraud like to hide — a subtle deviation easy to miss in a spreadsheet, a switched service level nobody flagged, a returns pattern that looks normal until it isn't.
DeepDetectAI's strength lies in its ability to surface these patterns with explainability, allowing operations and finance teams to act on exceptions before they accumulate. The system's integration into existing workflows and its provision of analyst support means companies can shift from after-the-fact reconciliation to real-time exception management.
As freight data continues to grow in complexity, the ability to deploy machine learning models that adapt to each shipper's unique profile — rather than relying on static rules — will become a competitive differentiator. The awards recognition from FreightWaves signals that the industry is beginning to separate genuine AI deployments from marketing buzz, placing a premium on systems that deliver measurable outcomes.
Sources:
