Amazon unveiled the next generation of its Proteus warehouse robot at its Delivering the Future event in London on Thursday, and the upgrade that matters isn't speed or payload — it's language. Workers can now assign the robot tasks "the way they'd communicate with a colleague," Gizmodo reports, with no programming interface or specialized commands in between.
"You tell it what needs to be done. It figures out the priority, the route, the timing," Amazon Robotics VP Scott Dresser said in the company's announcement. "It becomes your assistant for material movement."
Proteus debuted in 2022 as Amazon's first fully autonomous mobile robot, hauling carts of up to roughly 880 pounds — but only in designated dock areas. The new version roams anywhere in a facility that items need moving, and it joins a fleet that already numbers more than one million robots across Amazon's network. It's being piloted in Amazon's labs now, with European deployment planned for the first half of 2027 as part of a €10 billion ($11.6 billion) fulfillment modernization that Tech Startups reports includes a pledge to add 25,000 European fulfillment jobs.
Why natural language is the unlock
Every autonomous mobile robot vendor can move a cart. What has kept AMR deployments slow and expensive is everything around the robot: integration with warehouse management systems, task configuration, exception handling, and training floor workers to direct the things. A robot that takes conversational instruction collapses much of that overhead. Tasking becomes something any associate can do on day one, and re-tasking — the bane of peak-season operations, when priorities shift hourly — stops requiring a technician.
That should worry, and motivate, the rest of the warehouse automation market. Locus, Zebra's Fetch line, and the WMS platforms they plug into now have a public benchmark: if the largest fulfillment operator on earth says plain-language robot tasking is the new interface, "requires integration services" starts sounding like a legacy pitch. Expect natural-language layers to show up in every vendor roadmap by NRF Big Show.
The labor subtext
The timing is not subtle. CNBC's coverage framed the reveal against a backdrop of continuing AI-driven workforce reductions across big tech, and Amazon's own internal projections — leaked to The New York Times last fall — suggested automation could let the company avoid hiring more than 160,000 U.S. workers by 2027, with an eventual goal of automating 75% of operations. Amazon disputes that those documents reflect company strategy, and the 25,000-job European pledge is clearly designed to preempt the headline. Both things can be true: absolute headcount can grow in expansion markets while robots quietly absorb the marginal hour everywhere else.
For U.S. retailers, the practical takeaway is timeline. Europe gets Proteus in early 2027; the U.S. network — where the robot was born — presumably won't wait long after. As we covered Saturday in our look at the May jobs report, retail hiring is already frozen. Robots that anyone can boss around in plain English won't unfreeze it. The cost of automating a fulfillment operation just dropped again — not because the hardware got cheaper, but because the humans no longer need a manual.