Last updated: 2026-04-25
Last updated: April 2026
Boston Dynamics was founded in 1992 by Marc Raibert, a robotics researcher who had spent years studying legged locomotion at MIT's Leg Laboratory. His co-founder was Nancy Cornelius, who became the company's first employee. Raibert's PhD research had focused on the fundamental mechanics of balancing on legs — how to make a machine hop, run, and recover from stumbles — and Boston Dynamics was, from the start, an attempt to take that academic work and build real machines with it.
For the first decade, the company worked largely on simulation software and military-funded research. The big break came in 2005 with BigDog, a quadruped pack mule funded by DARPA. BigDog could carry 150 kg across rough terrain, and videos of soldiers kicking it while it stayed upright became some of the earliest robotics clips to go viral online. But BigDog was hydraulic and extremely loud — too loud for practical military use, as it turned out. The US Marine Corps eventually shelved the project.
Boston Dynamics kept building quadrupeds. LittleDog (2010) was a smaller research testbed. Cheetah (2012) set a land speed record for legged robots at 28 mph on a treadmill, though it was tethered and impractical. None of these were products. They were research platforms, each one feeding data and design insights into the next.
The real shift came in 2016 with SpotMini. Unlike its predecessors, SpotMini was entirely electric — no hydraulics, no engine noise. At 25 kg it was small enough to work indoors. It had a robotic arm for basic manipulation. And it looked, for the first time, like something you could imagine actually deploying in a building rather than hauling across a military test range.
A November 2017 redesign gave SpotMini its now-recognisable yellow shell and much smoother movement. A video of it opening a door went viral, and Boston Dynamics suddenly had a product people outside the robotics community were paying attention to.
Boston Dynamics' ownership history is unusual. Google's parent company Alphabet acquired it in December 2013 as part of a broader robotics shopping spree. But Google reportedly struggled to find a commercial direction for the company, and in June 2017 SoftBank Group bought it for an undisclosed amount.
Under SoftBank, the push toward commercialisation accelerated. Marc Raibert stepped back from the CEO role in 2019, handing over to Robert Playter, a long-time engineering leader at the company. Then in December 2020, Hyundai Motor Group agreed to acquire an 80% stake for $880 million, with the deal closing in June 2021. SoftBank retained 20%.
Each transition brought different resources. Google offered AI talent, SoftBank provided capital and patience, and Hyundai brought manufacturing scale. Through it all, the core engineering team remained largely intact — one of the reasons the company's technical output stayed consistent even as the business card on the front door kept changing.
The commercial Spot launched in September 2019 with an early-access leasing programme. Around 75 units went out to select customers — construction firms, energy companies, academic labs. In June 2020, Boston Dynamics opened sales to the public at $74,500 for the base Explorer kit. That price has not changed since.
Spot found its niche fast. Oil and gas companies used it to patrol refineries and offshore platforms. Construction firms sent it to document building progress. Power utilities deployed it inside substations. The common thread was always the same: Spot went where it was dangerous, boring, or impractical to send a person on a regular basis.
The COVID-19 pandemic gave Spot an unexpected early showcase. In March 2020, Brigham and Women's Hospital in Boston used one to carry a tablet and two-way radio through triage tents so doctors could screen patients without direct contact. It was a small deployment, but it put Spot on television at a moment when the public was unusually receptive to the idea of robots doing work that exposed humans to risk.
In February 2021, Boston Dynamics released the Spot Arm — a 7-degree-of-freedom attachment that turned Spot from a mobile sensor platform into something that could physically interact with its environment. The arm can open doors, turn valves, pick up objects, and drag items. At roughly $65,000 on top of the base robot, it was not an impulse purchase, but it significantly expanded what Spot could do on autonomous missions.
Software mattered as much as hardware. GraphNav, Spot's autonomous navigation system, lets the robot map a facility once and then repeat routes without human input. Orbit, the fleet management platform, lets operators schedule missions, review collected data, and manage multiple robots from a central dashboard. By 2022, Boston Dynamics reported customers were automating over a million data captures per year.
The AI layer keeps getting thicker. In late 2024, foundation model updates improved Spot's ability to recognise environmental hazards. By early 2026, an integration with Google DeepMind's Gemini Robotics ER 1.6 gave the robot reasoning capabilities for inspection tasks — deciding what to examine rather than just following a script.
As of early 2026, Boston Dynamics says more than 1,500 Spot units are deployed across over 40 countries. The robot has evolved from a research curiosity into a standard tool for a specific class of industrial work. It faces competition — ANYbotics' ANYmal, Unitree's quadrupeds, and a growing field of Chinese-built alternatives — but Spot has the advantage of years of real-world deployment data, a mature software ecosystem, and the manufacturing muscle of Hyundai behind it.
Spot Cam 2, released in January 2026, added 4K video with 25× optical zoom and integrated thermal imaging, pushing the robot further into the security monitoring space. Data centre operators, a newer market for Spot, report ROI within 18 to 24 months for perimeter patrol and thermal anomaly detection.
It is still not a mass-market product. At $74,500 for the base unit and well over $100,000 fully kitted out, Spot is a capital expenditure that makes sense only where the cost of not inspecting — an undetected gas leak, a missed structural crack, a delayed construction schedule — justifies the investment.
Some content on this page was created with the assistance of AI tools.