Subscribe to Our Newsletter

Success! Now Check Your Email

To complete Subscribe, click the confirmation link in your inbox. If it doesn’t arrive within 3 minutes, check your spam folder.

Ok, Thanks

Amazon acquires River Robotics to put legged delivery bots on your doorstep

The e-commerce giant is betting that four-legged autonomous robots can work alongside human drivers, not replace them. Whether that promise holds is another matter

Ian Lyall profile image
by Ian Lyall
Amazon acquires River Robotics to put legged delivery bots on your doorstep

Amazon has acquired River, a robotics startup that makes autonomous last-mile delivery robots, in its latest move to automate the final, most expensive leg of its logistics chain.

River's machines are distinctive: legged robots with wheeled feet and a long cargo box mounted on top, designed to handle the kind of terrain that wheeled-only robots struggle with, including steps, kerbs, and narrow pathways. The company had previously partnered with Just Eat Takeaway and VHO before catching Amazon's attention.

The intended use case is cooperative rather than solitary. A human driver drops a parcel at one address while the River robot simultaneously handles a delivery nearby, compressing the time each stop takes without removing the driver from the equation entirely.

The workforce question Amazon cannot dodge

Amazon has been careful in how it frames the deal. Its public statement stresses that the robots are built to work with delivery drivers, a pointed choice of words from a company that has spent years fielding criticism over automation and working conditions.

The sensitivity is understandable. Amazon's workforce is vast, and its relationship with delivery labour is already complicated by its contracting model, through which many drivers are employed by third-party logistics companies rather than Amazon directly. That structure limits Amazon's direct exposure to headcount decisions, but it does not insulate the company from public scrutiny when robots begin taking on tasks those contractors currently perform.

The honest answer is that nobody knows yet whether cooperative delivery will quietly erode driver numbers over time. Amazon's assurances are plausible in the short term. They are harder to bank on over a decade.

A decade in robotics, and still accelerating

Amazon's acquisition of River fits a pattern the company has been building since the early 2010s. It purchased Kiva Systems in 2012 and spent the following years deploying warehouse robots at scale, largely out of public view. What looked like a back-office efficiency play has since become one of the most sophisticated industrial robotics operations in the world.

Its external bets have ranged widely. Amazon invested in Rivian, the electric van maker now supplying its delivery fleet, and has backed Zoox, a self-driving robotaxi company, signalling that its interest in autonomous logistics extends well beyond warehouses.

River adds a missing piece: the doorstep itself. Warehouse automation is solved. Pavement-level, door-to-door delivery remains stubbornly human. River is Amazon's attempt to change that.

Why Amazon plays robotics differently

The comparison that keeps surfacing is Tesla, which is developing its Optimus humanoid robot partly as an enterprise product it intends to sell to other businesses. Amazon is not doing that, at least not yet.

Its approach is to build robotics capability in service of its existing businesses rather than as a standalone product category. The robots going into its fulfilment centres are not for sale to competitors. The delivery bots acquired through River will run on Amazon's routes, not a rival's.

The parallel that matters most is AWS. Amazon built its cloud computing infrastructure to serve its own e-commerce operation, then recognised it had something other businesses would pay for and turned it into a separate, enormously profitable division. Its robotics capability is not there yet, but the structural conditions for a similar move are beginning to take shape.

What to watch next

River's robots will not transform Amazon's delivery network overnight. Deployment at meaningful scale takes years, and last-mile robotics faces real-world obstacles, from weather to theft to the sheer variability of residential environments, that controlled demos do not capture.

But Amazon has a track record of absorbing promising robotics companies and building them into something larger and more durable than their founders imagined. River is unlikely to be the exception.

The more interesting question is not whether these robots work. It is what Amazon does once they do.

Ian Lyall profile image
by Ian Lyall