On April 15, Skild AI announced the acquisition of Zebra Technologies’ Robotics Automation business. The deal hands Skild control of the former Fetch Robotics fleet — one of the most deployed collections of autonomous mobile robots (AMRs) in commercial warehousing — along with Zebra’s Symmetry Fulfillment orchestration platform, which coordinates human and robotic workers across warehouse floors.

Zebra originally paid $290 million for Fetch Robotics in 2021. Financial terms of the sale were not disclosed.

Skild AI is the startup behind the “Skild Brain,” an omni-bodied AI model designed to control any robot type — humanoids, robotic dogs, tabletop arms, AMRs — without needing to be retrained for each body form. The company raised $1.4 billion in a Series C round in January 2026 and is now valued at more than $14 billion.

Deepak Pathak, CEO of Skild AI, described the strategic intent: “By combining Zebra’s human-robot orchestration platform with Skild AI’s omnibodied brain, we are set to transform what end-to-end automation looks like in warehouses that exist today. Zebra’s orchestration layer brings humans into the fold alongside Skild AI’s vision of ‘any robot, any task, one brain’ — turning warehouses into living symphonies of human and machine autonomy.”

What Skild Actually Built

The Skild Brain is not a warehouse-specific model. It is a foundation model for physical intelligence — trained to reason about robot bodies in general rather than individual hardware configurations. This matters because the robots deployed in a real warehouse are rarely all the same make and model. A facility might run AMRs from one vendor, picking arms from another, and humanoids from a third. Historically, each required its own software stack and often its own team.

Skild’s claim is that a single AI brain can run all of them simultaneously. The Zebra acquisition gives that claim an immediate test at scale. Zebra’s Symmetry platform already manages thousands of devices across hundreds of warehouses. Skild inherits that installed base.

The result, as Abhinav Gupta, President of Skild AI, put it: “The Skild Brain in conjunction with Zebra’s tried and tested person-to-goods solution will turn warehouses into hubs of hyper-efficiency.”

Why Zebra Sold

For Zebra, the sale represents a strategic refocus. The company entered robotics aggressively when it bought Fetch in 2021, but the standalone AMR market proved harder to grow as a hardware business than anticipated. Zebra’s core strengths lie in enterprise supply chain intelligence: RFID tagging, barcode scanning, mobile computers, and the software that connects all of it.

By selling the robotics arm to Skild, Zebra exits a hardware race it was not winning and redeploys capital toward the software and intelligence layers where it holds durable competitive advantage. It is a clean move.

The Bigger Picture: Physical AI Is Following the Same Playbook as Software AI

For the past three years, the enterprise AI story has largely been a software story. AI agents taking over email management, customer service queues, data pipelines, and finance workflows. The pattern has been consistent: identify a repetitive, high-volume task, wrap an AI agent around it, and let it run.

The Skild acquisition shows the same pattern arriving in physical operations, just about two years behind the software curve.

The underlying logic is identical. Rather than dozens of disconnected robotic systems, each with its own controller, you get one intelligent layer that orchestrates them all. Rather than programming each arm or robot explicitly for each task, you train a generalised AI that can adapt to whatever body and environment it finds itself in.

This is what “agentic AI” looks like in a warehouse. One orchestration layer. Multiple specialised agents (some robotic, some software). Unified by a common intelligence.

The same shift is well underway in digital operations. Platforms that unify AI agents across departments are replacing point solutions, for the same reasons Skild is replacing per-robot software stacks.

What This Means for Business

Most business owners do not run warehouses. But the implications of this deal stretch beyond logistics.

The consolidation of physical and digital AI is accelerating. The companies deploying the most effective AI operations in 2026 are the ones that have moved from a portfolio of individual tools to a unified coordination layer. That is true in software and it is now becoming true in physical environments. If your business has operational components in either domain, the question is not whether this matters but when.

Operations teams should be auditing their own fragmentation. If your business relies on multiple separate systems — whether software platforms or physical equipment — that do not share intelligence or data, you are running the pre-Skild warehouse model. Every time your people have to manually move information between systems or re-brief a tool from scratch, there is a coordination failure that AI should be filling.

The “one brain” model is the right intuition, but implementation is the hard part. Skild’s value proposition is elegant: a single AI model that understands any robot. The implementation challenge is real — combining Zebra’s orchestration layer with a general-purpose robotics foundation model is technically non-trivial. Businesses evaluating similar unified AI platforms in software should apply the same scrutiny: does the vendor have the integration depth to actually coordinate your existing systems, or are they selling the concept before the product exists?

For manufacturing, logistics, and retail businesses: the physical AI wave is no longer theoretical. Skild now has the robots, the orchestration platform, the capital, and the installed customer base to move fast. If your sector competes on operational throughput, this is a supply chain story worth watching closely.

---

If you’re deciding where to start with agents, start here. The free Working With Claude field guide walks through the ecosystem, Claude Code, and a real rollout plan. Get your copy.