Agility Robotics helped shape the modern humanoid industry before most of its competitors even existed, but now it's struggling to keep pace with what it helped create. The Oregon-based robotics firm got its AI-powered, two-legged machine into the Guinness World Records years before China's humanoids started going viral for speed. Its influence can be seen in today's robots that are smashing human records. Despite its pioneer status, Agility is heading into what could be its harshest stress test yet: the public markets.

How Did Agility Robotics Become a Humanoid Pioneer?

Agility Robotics grew out of a research project that began around 2008 at Oregon State University, where Jonathan Hurst established the Dynamic Robotics Laboratory after earning his PhD from Carnegie Mellon University. Similar to MIT's Leg Laboratory that spawned Boston Dynamics, the program sought to build walking machines that could navigate the world as efficiently as people and animals.

One of the lab's early creations was known as ATRIAS, an acronym for "Assume The Robot Is A Sphere." The team built ATRIAS to test a simplified model of walking that treated the body like a spring-supported mass. Though its aesthetics were very unhuman-like, ATRIAS could move more naturally and efficiently than leading humanoids of the era like Honda's ASIMO. The breakthrough established the design philosophy that Agility would later build upon.

Hurst formed Agility Robotics as a university spinout in 2015 with Damion Shelton and Mikhail Jones. Shelton brought business experience from cofounding threeRivers 3D, a Pittsburgh-area 3D-imaging startup. Jones had worked closely with Hurst as a graduate student and contributed to the control systems behind the lab's robots. They secured almost $800,000 in early funding from The Robotics Hub, a CMU-linked startup accelerator.

What Made Cassie and Digit Breakthrough Moments?

Agility's founding team had its first major win in 2016 when it secured a million-dollar DARPA contract to turn its work from ATRIAS into a more capable robot named Cassie. Introduced in 2017, Cassie had a pair of robotic legs with a compact torso and exposed mechanical joints. At about 31 kilograms (68 pounds), the robot was half the weight of ATRIAS.

Cassie became famous for its running capabilities. In 2021, the biped finished a 5-kilometer run in about 53 minutes on a single battery charge. In 2022, a Cassie robot set a Guinness World Record for the fastest 100-meter sprint by a two-legged robot, finishing in 24.7 seconds. Though slower than human runners, Cassie completed both feats autonomously using machine learning control, with its digital twin trained through thousands of simulated runs before transferring what it learned to the real robot.

Cassie's successor brought Agility a new level of attention. By the time the company introduced its flagship robot worker, Digit, in February 2019, it had raised nearly $9 million to date. The original Digit stood about 155 centimeters (5 feet 1 inch) and weighed 42 kilograms (93 pounds). It inherited Cassie's proven leg design but added a larger torso, arms, sensors, and much more onboard computing. It could carry and stack boxes weighing up to 18 kilograms (40 pounds).

Ford became the first commercial buyer in 2020, taking two Digit robots to explore autonomous last-mile delivery. Ford envisioned a delivery system where Digit would ride in a self-driving van, unfold itself, and carry a package to the recipient's door. The concept grabbed headlines, but the automaker quietly abandoned the project as Agility shifted its focus toward warehouse and factory work.

Digit had its biggest redesign in 2023 after Agility secured around $150 million in a Series B round the prior year, led by DCVC and Playground Global. Amazon also participated through its billion-dollar Industrial Innovation Fund. The new Digit had a much larger and more industrial build than the first iteration, standing 175 centimeters (5 feet 9 inches).

What Separates Real Deployment From Impressive Demos?

As humanoid robotics accelerates, the language around it is starting to blur. A robot walks across a stage, folds a shirt, carries a tote, or completes a task after a spoken instruction. The video is real. The engineering may be excellent. But none of those facts, by themselves, tell us whether the robot is commercially ready.

The central distinction is this: a demo proves possibility, while a deployment must prove repeatability, context, and operational consequence. A polished demonstration can travel around the world before anyone has answered basic operational questions about where the task happened, how many hours the robot worked, how often a person intervened, whether the customer was paying, and whether the machine returned the next day to do the same job again.

At Physical AI Radar, experts have found it useful to separate three things that are often collapsed into one headline: demonstrated capability, the quality of the evidence behind that capability, and commercial reality. A robot can score highly on one dimension and remain uncertain on another.

Steps to Evaluate Humanoid Robot Maturity

• Manufacturer Demonstration: The company shows the robot performing a task, which establishes possibility under the demonstrated conditions but does not prove broader capability.
• Testing or Pilot in Real Customer Environment: The robot operates in a factory or warehouse, introducing variability, safety requirements, production constraints, and consequences for failure, but a pilot is still a test whose purpose may be to discover what does not work.
• Contracted Deployment: A named customer, commercial agreement, defined workflow, or paid service relationship strengthens the evidence considerably because another organization is committing resources to the system.
• Recurring Work With Measurable Results: The strongest public evidence arrives when recurring work can be tied to measurable operating results, including a defined task, a real site, sustained runtime or duration, throughput, repeated shifts, and customer confirmation.

Consider Agility Robotics' Digit as an example. GXO announced in June 2024 that it had signed a multi-year agreement with Agility Robotics to deploy Digit under a Robots-as-a-Service model at GXO's facility in Flowery Branch, Georgia. Agility later reported that Digit had moved more than 100,000 totes at that facility. The significance comes from recurrence; repeating a useful task in a live logistics workflow is closer to the kind of evidence an operator cares about.

However, a vendor-reported milestone is not identical to a complete independent operating audit. Throughput totals are valuable, but they are even more valuable when accompanied by denominators such as fleet size, scheduled hours, and intervention rates.

Why Is Agility's Public Market Push So Risky?

Despite its pioneer status and the opening of what it calls the world's first dedicated humanoid robot factory, Agility is struggling to keep pace with what it helped create. China is scaling much faster, and Western rivals are raising more money and dominating the hype cycle. Now Agility is heading into what could be its harshest stress test yet: the public markets. Its plan is already divisive, and the disclosures behind it show just how much pressure is building beneath the surface.

As Agility finds itself in its biggest spotlight yet, the fundamental question remains: can being early still matter? Or will it become a cautionary tale of the physical AI boom? The answer will likely depend on whether Agility can translate its technical achievements into the kind of sustained, measurable commercial deployments that separate genuine products from impressive experiments.