This Humanoid Robotics Company Is Going Public, but Its CEO Warns Not to Expect a Home Robot Anytime Soon

The humanoid robotics industry is experiencing significant financial success. Recently, AI2 Robotics, a Shenzhen-based startup developing wheeled humanoid robots, acquired nearly $735 million, resulting in a valuation of approximately $3 billion. Earlier this year, Apptronik, an Austin firm focusing on humanoid robots for manufacturing and logistics, closed a funding round of $935 million, bringing its valuation to over $5.5 billion, with support from heavyweights like Google, Mercedes-Benz, and John Deere. Furthermore, last fall, Figure AI, a San Jose startup specializing in general-purpose humanoid robots, revealed it had secured $1 billion in Series C funding, boasting an eye-watering valuation of $39 billion.

In contrast, Peggy Johnson, CEO of Agility Robotics, takes a cautious stance. We spoke by phone last week shortly after the company revealed its plan to go public through a merger with Michael Klein’s Churchill Capital Corp XI, a special purpose acquisition company (SPAC). This deal places Agility’s valuation at roughly $2.5 billion and is projected to yield over $620 million in gross proceeds—the largest capital raise in humanoid robotics to date. The merger awaits shareholder approval and SEC review, expected to be completed later this year.

Founded in 2015 as a spinoff from Oregon State University and based in Salem, Oregon, Agility manufactures bipedal humanoid robots designed for warehouse and factory usage. This move is notable for several reasons. It positions Agility as the first dedicated humanoid robotics firm to go public, allowing retail investors access to a domain previously controlled by affluent VC funds. Additionally, it offers a rare look into a sector where competitors usually keep financial details and technological advancements closely guarded.

Johnson, who has previously held significant positions like executive vice president of business development at Microsoft, where she played a part in the $26 billion acquisition of LinkedIn, and later as CEO of Magic Leap, guided our conversation with deliberation. She declined to provide future financial forecasts, withheld details on the bill of materials for Agility’s flagship robot, Digit, and adeptly avoided speculative questions.

When asked why Agility opted for a SPAC route rather than another private funding round—thereby bypassing the traditional IPO roadshow and its associated pricing pressures—Johnson explained that Agility’s pioneering position in this public space granted them a first-mover advantage. For investors keen on acquiring shares of an innovative robotics company, Agility represents “an acceleration story and a timing story,” she articulated. The raised capital will also be aimed at boosting production capabilities at the 70,000-square-foot manufacturing facility in Salem and fulfilling existing customer orders.

Addressing the negative perceptions surrounding SPACs—many companies that chose this path in 2021 saw notable declines or even failures—Johnson remained unfazed. “If we stay focused, delivering customer by customer, robot by robot, we aspire to sidestep similar volatility,” she remarked. “Our primary competitor at present is ourselves—how quickly we can perform and broaden our capabilities.”

The pipeline extends far beyond pilot projects, Johnson shared with TechCrunch, revealing over $300 million in booked, multi-year revenue consisting of around 1,000 robots operating under a robots-as-a-service model, where customers pay a monthly fee instead of buying the machines outright. “Everyone on our list right now has been meticulously vetted and has deployment plans supporting their proof of concepts,” Johnson noted, mentioning clients like GXO Logistics, Amazon, Toyota Motor Manufacturing Canada, Schaeffler, and Mercado Libre.

Digit, Agility’s flagship robot, is intentionally designed for simplicity. Standing at approximately 5’9″ and weighing about 160 pounds, it excels at moving heavy objects within human-created environments. Its standout feature is its reverse-bend knees, which resemble “bird legs,” allowing it to reach from floor level to overhead shelves without encountering knee collisions with warehouse storage. Johnson explained that the founders of Agility were not pursuing biomimicry for its own sake; the robot’s hands, with two thumbs and two fingers, are specifically designed for gripping heavy plastic totes, even when their contents shift during transit.

Johnson stated that Agility is “LLM-agnostic,” using models such as Claude and Gemini to manage what she calls the semantic layer—transforming high-level instructions into robotic actions. She recounted a recent test where engineers scattered various trash types on the floor and instructed Digit to “clean up this mess.” The robot effectively assessed, sorted, and collected all items accurately, even identifying bubble wrap as non-recyclable.

However, Agility’s true strength lies in the physical layer—mechanics of balance, locomotion, and manipulation—an advantage gained from over a decade of real-world deployment. “While LLMs had access to the entire internet for training,” she said, “the physical AI for humanoids is still being developed.” Johnson believes Agility stands out: “We might possess the largest data lake of operational robotics data from genuine real-world scenarios.”

Besides raw data, Johnson emphasized safety as a critical differentiator between Agility and its rivals. While competing firms often showcase their robots in controlled demos and polished videos, Agility must adhere to actual industrial safety certification standards to function within customer facilities. “You can’t just build a robot and then ensure its safety,” she remarked. “That would necessitate redesign. All safety elements—the electrical system, every component, and the associated software—must be certified from the outset.” (This concern is especially relevant given the potential presence of humans in the workplace. In November, Figure AI’s former head of product safety filed a lawsuit against the company, claiming he was dismissed after raising fears that its robots might cause serious injuries. Figure contests these claims.)

As for home integration, Johnson believes humanoid robots will eventually find their place there, but cautions against expecting features like breakfast delivery in the near future. She estimates it will take “10-plus years” for this to come to fruition, noting that while warehouses and factories, despite their complexities, have fixed layouts and predictable operations, homes are often chaotic, filled with pets, children, visitors, and assorted objects that can be strewn about.

“At least roads have some structure to them,” Johnson stated, comparing the challenge to that of autonomous vehicles. “Unfortunately, many environments where humanoids will work lack that same organization.”

Agility is not overlooking the home market. Johnson mentioned that the company will explore it when the timing is right. For now, however, their focus remains intensely on the warehouse sector, particularly in light of the increasing numbers of retiring workers and the younger generations’ reluctance to undertake physically demanding jobs. “There are currently over a million unfilled positions in the U.S. in these sectors,” she emphasized. “These roles are incredibly tough to fill.”