Pallet's New AI Agents Bring Autonomous Action to Legacy Supply Chains

SAN FRANCISCO – October 20, 2026 — For the past decade, the holy grail of global logistics was visibility. Billions of venture capital dollars poured into building digital "control towers"—massive software platforms designed to track every shipping container, pallet, and truck traversing the globe. We achieved the visibility, but in doing so, we uncovered a far more complex bottleneck: knowing that a typhoon is about to shut down a critical port in Southeast Asia does not automatically reroute a fleet of ships, adjust warehouse staffing in Ohio, or notify waiting customers. Human operators still have to manually process the disruption across a fragmented web of legacy software.

Today, Pallet, a San Francisco-based artificial intelligence company, announced a new capability called Signals, aiming to close the gap between knowing and doing. Signals is designed to inject live external events—shifting tariffs, raw material shortages, port congestion, and extreme weather—directly into enterprise AI agent decision-making. By linking the chaos of the outside world with an enterprise’s operational data, the platform allows autonomous agents to execute complex, multi-step responses across supply chain networks without human intervention.

For the physical economy, this represents a structural shift in how business is conducted. It marks the transition from passive tracking dashboards to predictive, agentic decision intelligence capable of executing operational changes across the massive, aging enterprise systems that actually power global trade.

From Visibility to Autonomous Action

The supply chain technology landscape has long been dominated by real-time transportation visibility platforms. Established players have mapped the world's freight networks, providing end-to-end tracking and predictive estimated times of arrival. Yet, the limitation of visibility platforms is that they stop at the alert. They are the fire alarm, not the sprinkler system.

Pallet’s approach with Signals shifts the paradigm toward agentic execution. The system expands upon the company's existing AI intelligence layer, Atlas, which maps the intricate operational relationships within a supply chain. When an external disruption occurs—say, an unexpected aluminum shortage—Signals traces the dependencies across the entire operation. It calculates the ripple effects on production schedules, customer orders, inventory levels, transportation capacity, and service commitments.

Crucially, it then formulates and executes a response. "Our customers don’t care why a shipment is late. They care that we told them early and already had a plan," said David Radom, Group CEO at Prism Logistix, an early adopter of the platform. "One port delay or storm on a key lane can ripple across our entire network. Pallet’s Signals technology shows us those ripples days before they reach a customer, so by the time we call, we've already rerouted and rescheduled. With Pallet, our conversations with customers are proactive instead of apologetic."

This capability is broken down into four core functions: ingesting external and enterprise context, providing native access across enterprise systems, reasoning across complex operational dependencies, and maintaining strict enterprise-owned intelligence and governance. The AI agents continuously update their reasoning as new information arrives, ensuring that the automated response evolves in tandem with shifting real-world conditions.

Bridging the Generative AI and Green Screen Divide

Perhaps the most significant, yet underappreciated, aspect of Pallet’s technology is its architectural approach to legacy systems. The modern digital economy might run on cloud-native application programming interfaces, but the physical economy still relies heavily on decades-old infrastructure. Massive logistics providers, warehouse operators, and global manufacturers continue to depend on AS/400 mainframes—the infamous "green screens" of the computing world.

Replacing these monolithic systems is a billion-dollar risk that most chief information officers are unwilling to take. They are incredibly robust, highly secure, and deeply entrenched in the daily workflows of millions of workers. The challenge has always been how to bring modern machine learning capabilities to these siloed, aging databases.

Pallet bypasses the need for expensive, multi-year digital transformation projects by integrating natively with these legacy tech stacks. The company's agents can interface with AS/400 mainframes, traditional enterprise resource planning software, transportation management systems, and warehouse management systems without requiring modern connection protocols or database migrations.

"The real bottleneck isn't knowing a shipment is late; it's the thirty manual keystrokes required across three different legacy portals to fix it," noted one logistics integration architect familiar with the system's deployment. By acting as a versatile orchestration layer, Pallet's agents can effectively use the software just as a human operator would, executing commands and updating records across disparate platforms in real-time.

"The physical economy runs on an enormous amount of knowledge and expertise, much of it still trapped in legacy systems and in people’s heads," said Sushanth Raman, Founder and CEO of Pallet. "AI has the potential to unlock that capacity and fundamentally change how the world produces and moves the food, products, and materials we rely on every day. Signals is another step toward making that possible."

Raman’s background—having previously led data science for logistics divisions at major tech firms—highlights the caliber of engineering being directed at these industrial challenges. The goal is not to replace the legacy systems, but to build an intelligent, autonomous layer on top of them.

The Economics of Enterprise-Owned Intelligence

The deployment of autonomous AI agents in mission-critical supply chains naturally raises questions about risk and control. If a generative AI model hallucinates a response in a customer service chatbot, the result is an embarrassing screenshot on social media. If an autonomous agent hallucinates a supply chain command, the result could be millions of dollars in misdirected freight or halted production lines.

Pallet addresses this through strict enterprise governance controls. The Signals platform keeps operational context, persistent memory, and the AI stack completely under the enterprise’s control. Organizations maintain granular governance over exactly what the agents are permitted to access, decide, and execute. A human-in-the-loop can be mandated for high-stakes decisions, while routine rerouting can be fully automated.

This pragmatic approach to AI deployment has resonated with major players in the physical economy. Pallet currently reports deployments across more than seventy logistics and manufacturing enterprises, including industry giants like Knight-Swift, Lineage, Penske, STG Logistics, and GXO. These are not pilot programs in innovation labs; these are deployments across asset-based carriers, freight forwarders, and massive distribution networks.

The rapid adoption is fueled by a strong syndicate of venture capital. Backed by over twenty-five million dollars in disclosed funding from prominent firms including General Catalyst, Bain Capital Ventures, Activant Capital, and Bessemer Venture Partners, Pallet has the resources to scale its engineering efforts and support massive enterprise integrations.

Structural Shifts in Global Competition

As global supply chains face an era of unprecedented volatility—driven by geopolitical fragmentation, climate change, and shifting trade policies—the ability to react instantaneously to external shocks is no longer a luxury; it is a baseline requirement for survival. We are moving from an era of strictly optimized manufacturing to resilient, adaptable networks, and AI agents are the infrastructure enabling that shift.

"Supply chains are the infrastructure behind everyday life," said Bart De Muynck, Founder at Better Supply Chains and former Chief Industry Officer at Project44. "AI that can understand how disruptions ripple through an operation and help companies before they escalate has the potential to change how the physical economy operates from the ground up."

The introduction of Pallet's Signals represents more than just a new software feature. It is a fundamental rewiring of how industrial data is processed and acted upon. By bridging the gap between the chaotic external world and the rigid, legacy systems of the physical economy, autonomous agents are poised to become the new nervous system of global trade. As these intelligent systems learn to navigate tariffs, weather events, and material shortages with superhuman speed, the companies that adopt them will secure a compounding operational advantage that will define the next fifty years of industrial progress.