News

The Rise of Amazon AI Robots

• By David Ramel
• 11/05/2025

Along with heavily investing in advanced AI to help its users and pertinent cloud developers create agents and other systems, Amazon is putting the tech to work in doing what the company does best: sell stuff.

Amazon is scaling an AI-robotics playbook across its operations, reporting more than 1 million robots deployed since 2012 and introducing new systems that target faster delivery and safer, less repetitive work in fulfillment centers. A recent look inside its network shows how established platforms like Hercules, Titan, Robin, Cardinal, Sparrow, Proteus, Sequoia, and packaging automation are now joined by newer additions, including Vulcan and the ceiling-mounted Blue Jay system coordinated with an AI decision-support model called Project Eluna. Together, the technologies highlight a push to accelerate order flow while keeping employees in more ergonomic power zones and away from repetitive reaching, lifting, and floor travel.

Blue Jay is designed as an all-in-one streamlined workspace for same-day operations. The system coordinates multiple robotic arms within a single workspace to pick, stow, and consolidate items that historically required several separate stations. Amazon describes it as handling tens of thousands of items at high speed and moving inventory through facilities with a juggler-like cadence. The concept-to-production cycle was shortened to just over a year by using digital twins and accumulated data from the existing fleet, and early production testing is underway at a site in South Carolina. In these tests, Blue Jay can already pick, stow, and consolidate a broad majority of item types and is positioned to become core infrastructure for Same-Day sites as it matures.

Project Eluna is framed as an agentic AI model that assists operations managers. Rather than juggling dozens of dashboards, managers can ask targeted questions such as where to shift people to avoid a bottleneck and receive data-backed recommendations. The pilot begins at a fulfillment center in Tennessee with an initial focus on sortation optimization, pulling on historical and live building data to anticipate problems before they cascade. The initiative reflects a broader theme in Amazon operations: use AI to compress decision loops, reduce firefighting, and let leaders spend more time coaching teams.

New systems build on a foundation that includes Vulcan, a robot with a sense of touch that focuses on the highest and lowest rows of inventory pods to keep employees working at mid-thigh to mid-chest height, and DeepFleet, a generative AI foundation model aimed at coordinating large fleets of mobile robots. In the broader network, Hercules and Titan drive units transport item pods; Sparrow picks individual items into totes using computer vision; Robin and Cardinal sort and place packages with robotic arms; Proteus moves autonomously in shared spaces using onboard sensing; and packaging automation produces right-sized, curbside-recyclable paper bags to reduce materials and plastic usage. Sequoia consolidates inventory to free space and speed transactions, and an advanced Shreveport facility showcases how multiple systems can be orchestrated together in a single building.

The throughline across these efforts is employee support. Systems like Vulcan and Blue Jay are meant to keep work inside safer, more neutral body positions and to limit repetitive movements. AI planning tools aim to lighten the cognitive load on supervisors. Amazon also pairs the technical changes with workforce training, apprenticeships in mechatronics and robotics, and new AI-focused learning so employees can work confidently with the tools on the floor.

For customers, the near-term impact is quicker movement of items through facilities and gains in same-day throughput. For employees, it is a growing set of AI-enabled systems designed to shoulder the most repetitive tasks while surfacing smarter, earlier recommendations for people who keep the network running. Amazon says this is less about replacing people and more about augmenting them with practical automation that works in close concert with human teams.

More on Amazon's robots inside fulfillment centers is available at this overview of deployed systems. The company also detailed the Blue Jay system and Project Eluna in a companion announcement.