Reuters Supply Chain USA 2026 brought roughly 500 senior executives from global retailers, manufacturers and logistics operators to Chicago on June 23 and 24. The conversation on the floor and on stage covered cost visibility in a volatile trade environment, AI that delivers operational results rather than roadmap slides, and building supply chains resilient enough to hold up when conditions shift.
The event's organizing theme, "Future-Proofing Supply Chains to Bend, Not Break," shaped four agenda pillars: designing for disruption, staying customer-focused, scaling innovation with measurable outcomes, and building partnerships tied to shared performance. What ran through all of it was a demand for specificity. Attendees weren't interested in frameworks. They wanted to know what was working in production, at scale.
The Key Themes From Reuters Supply Chain USA 2026
The gap between AI investment and AI output was the most consistent frustration in the room. Organizations have spent the past two years standing up pilots, and many are still waiting for those pilots to become operational programs. The driving question wasn't whether to invest in AI — it was how to move from experimentation to measurable impact without losing organizational momentum or trust.
Data readiness surfaced repeatedly as the core blocker behind that gap. The organizations making meaningful progress with AI were, almost without exception, the ones that had done the unglamorous work first: normalizing data across systems, establishing a single source of truth, and building governance structures that gave decision-makers confidence in what the data was actually telling them. The point that landed hardest across multiple sessions: AI applied to bad data produces bad decisions faster.
Workforce adaptation was the third recurring thread. The question of what AI does to jobs is present across every industry, but it felt particularly urgent in logistics, where significant portions of existing workflows are candidates for automation. The more grounded perspective from leaders in the room was that the goal isn't replacement — it's redeployment. Organizations with clear answers to "what do our people do next?" were the ones with sustainable AI programs.
Blake Tablak at Reuters Supply Chain USA 2026: Scaling AI from Hype to Operational Impact
On day one, Trax CEO Blake Tablak joined a panel session titled "Scaling AI in Logistics: From Hype to Measurable, Operational Impact," alongside Ken Ogada, Executive Director and Head of Americas Distribution at AstraZeneca, and Gregory Javor, SVP of Global Supply Chain Operations at Mattel. Ninaad Acharya, CEO of FulfillmentIQ and host of the eCom Logistics Podcast, moderated.
Acharya opened with a challenge most in the room recognized immediately: AI initiatives across the industry are failing to exit the pilot stage at a meaningful rate. The panel's collective answer was that the culprit isn't the technology — it's the conditions the technology gets dropped into.
Why Freight Audit AI Requires a Data Foundation First
Tablak spoke to this from Trax's position in the freight audit and payment space. Freight data has historically been fragmented, non-normalized and difficult to act on across carriers, modes and geographies. The work Trax has done at the data layer — normalizing invoice data across 21,000+ global carriers and 300+ data concepts — is what makes AI applications in freight audit actually function in production.
AI that surfaces cost savings recommendations or flags exception patterns only generates value when the underlying data is clean and consistent enough to trust. That's not a technology problem. It's a data infrastructure problem that has to be solved before AI can do anything useful with it.
He also addressed workforce directly. As AI automates more of the transactional work in freight audit and payment processing, Trax's approach is to retrain employees for higher-value roles rather than reduce headcount. The organizations that will get the most from AI are the ones that invest in their people's ability to work alongside it — not the ones that treat automation as a cost-reduction exercise alone.
What AstraZeneca and Mattel Are Doing With Supply Chain AI
The other panelists brought concrete perspectives from large-scale deployments.
Ken Ogada described AstraZeneca's "self-healing supply chain" initiative, which uses AI to monitor transit lead times across approximately 2,000 lanes globally and issue real-time routing recommendations. AstraZeneca built a governance layer around that system to ensure human oversight of automated decisions in high-risk processes — a detail that drew significant audience interest. The company also runs a mandatory AI training program across all employee levels, from foundational to advanced, to build organizational AI capability rather than concentrate it in a small technical team.
Gregory Javor walked through Mattel's deployment of autonomous mobile robots in value-added services operations, and the company's success cutting carrier fines from $30 million to $3 million through a combination of automation and better data contextualization. The challenge Mattel is actively working through — making sense of data spread across disparate platforms — resonated with most supply chain finance and operations leaders in the room.
The session closed with a consistent message from all three panelists: don't wait for perfect conditions. Identify the specific operational question your organization needs answered, find the data that can answer it, and build from there. Broad AI transformation programs tend to stall. Focused, data-grounded use cases tend to ship..png?width=718&height=205&name=May%20(1).png)
What This Means for Supply Chain and Finance Leaders
The organizations separating themselves right now aren't necessarily the ones with the most sophisticated AI. They're the ones that have done the foundational data work that makes AI trustworthy, and the cultural work that makes it actually get adopted.
For supply chain and finance leaders managing global freight spend, that means the conversation about AI has to start earlier in the stack — at the data layer, before it reaches the model layer. Normalized, enriched, audit-grade transportation data isn't just an input to freight audit. It's the foundation for every AI use case that follows: exception automation, procurement analytics, carrier benchmarking, cost allocation at the SKU level, and Scope 3 emissions reporting.
That's the work Trax has been doing for over 25 years. It was good to be in a room where more of the industry is catching up to why it matters.
Curious how Trax approaches AI in freight audit and payment? Talk to our team.
Also worth reading: How Finance and Logistics Leaders Use AI to Reduce Transportation Spend — a recap of Blake Tablak's session at the American Supply Chain Summit.