The humanoid robotics market is awash in cash proper now. Final week, AI2 Robotics, a Shenzhen-based startup that makes wheeled humanoid robots, raised roughly $735 million at an almost $3 billion valuation. Earlier this yr, Apptronik, an Austin-based maker of humanoid robots for manufacturing and logistics, closed a $935 million funding spherical valuing the corporate at greater than $5.5 billion. Final fall, Determine AI, a San Jose-based startup creating general-purpose humanoid robots, self-reported that it closed on $1 billion in Sequence C funding at an eye-popping $39 billion valuation.
By comparability, Peggy Johnson, CEO of Agility Robotics, is surprisingly measured. We spoke by telephone final week, simply after the corporate introduced plans to go public by means of a merger with Michael Klein’s Churchill Capital Corp XI, a particular goal acquisition firm, or SPAC. The deal values Agility at round $2.5 billion and is predicted to boost greater than $620 million in gross proceeds, the biggest capital elevate in humanoid robotics historical past. It hasn’t closed but; the merger nonetheless wants shareholder approval and SEC evaluation, and is predicted to be accomplished later this yr.
Agility was based in 2015 as a derivative from Oregon State College. Primarily based in Salem, Oregon, the corporate makes bipedal humanoid robots designed to work in warehouses and factories. Its SPAC maneuver is notable for a number of causes. It could make Agility the primary pure-play humanoid robotics firm to commerce on public markets, giving retail traders direct publicity to a sector that has up to now been out there primarily to deep-pocketed VC funds. It additionally presents a uncommon window into the funds of a enterprise in an area the place most rivals carefully guard their numbers and even the state of the tech they’re constructing.
Johnson — previously govt vp of enterprise growth at Microsoft, the place she helped engineer the $26 billion acquisition of LinkedIn, and later CEO of Magic Leap, the once-hyped augmented actuality headset maker — was cautious all through our dialog. She declined to supply forward-looking monetary steering, declined to reveal the invoice of supplies for Agility’s flagship robotic Digit, and pushed again politely at any time when questions veered towards hypothesis.
Requested why Agility goes public by way of a SPAC relatively than elevating one other personal spherical — a construction that skips the roadshow and pricing scrutiny of a conventional IPO — Johnson mentioned a lot of it boils all the way down to the first-mover benefit the corporate enjoys when it’s the primary of its ilk to go public. For traders clamoring for shares in a buzzy robotics firm, Agility is “an acceleration story and a timing story,” she mentioned. The proceeds may also assist Agility ramp up manufacturing at its 70,000-square-foot manufacturing facility in Salem, Oregon, and fulfill an present pipeline of buyer orders.
As for the troubled status of SPACs — many corporations that went public that approach in 2021 famously fizzled out totally or commerce nicely beneath their providing value — Johnson was unfazed. “If we simply maintain our head down, maintain delivering buyer by buyer, robotic by robotic, we hopefully gained’t expertise the identical volatility,” she mentioned. “Our greatest competitor proper now could be simply us. How rapidly we will execute, how rapidly we will proceed so as to add new expertise.”
The pipeline goes nicely past pilots, Johnson instructed TechCrunch, pointing to greater than $300 million in booked, multi-year income that represents roughly 1,000 robots which can be a part of a robots-as-a-service mannequin during which prospects pay a month-to-month price relatively than buying the machines outright. “All people on our checklist proper now could be already vetted, and so they have deployment plans behind their proof of ideas,” Johnson mentioned. Clients embrace GXO Logistics, Amazon, Toyota Motor Manufacturing Canada, Schaeffler, and Mercado Libre.
Digit itself is a intentionally unfussy piece of {hardware}. It stands about 5’9″, weighs round 160 kilos, and is designed to do one factor exceptionally nicely, which is transfer heavy objects in human-built areas. Its most distinctive characteristic is a set of reverse-bend knees — they’ve been referred to as “fowl legs” — that permit it to succeed in from ground degree to overhead shelving with out the knees colliding with warehouse racking. (Agility’s founders, Johnson defined, weren’t taken with biomimicry for its personal sake.) The robotic’s palms — two thumbs and two fingers — are equally task-specific; they’re optimized for gripping heavy plastic totes, at the same time as their contents shift in transit.
Johnson mentioned Agility is “LLM-agnostic,” drawing on fashions together with Claude and Gemini to deal with what she calls the semantic layer — translating high-level directions into robotic conduct. She described a latest take a look at during which engineers scattered several types of trash on the ground and instructed Digit merely to “clear up this mess.” The robotic assessed, sorted, and binned every thing appropriately, together with appropriately figuring out bubble wrap as non-recyclable.
After all, it’s the bodily layer — the mechanics of steadiness, locomotion, and manipulation — that Agility considers its core proprietary benefit, one constructed up over greater than a decade of real-world deployment. “The LLMs had all the web to coach on,” she mentioned. “When you concentrate on the bodily AI of humanoids — that doesn’t fairly exist but.” At most corporations, anyway. Johnson believes Agility is the exception: “We could have the biggest information lake of precise working robotics information in real-world environments.”
Past uncooked information, Johnson mentioned, security is the place the gulf between Agility and its rivals is greatest and most consequential. Whereas rival corporations showcase their robots in lab demos and choreographed videos, Agility has needed to meet precise industrial security certification necessities to function inside buyer services. “You may’t construct your robotic after which make it secure,” she mentioned. “That’s a redesign. You need to have all the security licensed — {the electrical} system, all the components, and the software program to help all of that.” (It’s not a trivial concern provided that people are sometimes someplace within the room. Again in November, Determine AI’s former head of product security sued the corporate, alleging he was fired after elevating issues that its robots had been highly effective sufficient to fracture a human skull. Determine has disputed the claims.)
As for the house, Johnson thinks humanoids will get there finally, however she mentioned to not anticipate them to ship breakfast in mattress anytime quickly. It’ll be “10-plus years,” she mentioned of the timeline, observing that warehouses and factories, for all their complexity, have fastened aisles and predictable tools and workflows not like houses which can be chaotic, with canines, infants, guests, and objects left in surprising locations.
“A minimum of roads have some self-discipline to them,” Johnson added, evaluating the problem to that of autonomous automobiles. “Many of the areas that humanoids will likely be working in don’t.”
Agility isn’t ruling out the house market. Johnson mentioned the corporate will enter it when it is sensible. For now, although, it’s laser centered on the warehouse market, given the rising numbers of retiring staff and youthful staff who aren’t prepared to take bodily demanding roles. “There’s one thing like over 1,000,000 jobs within the U.S. right now in these areas which can be unfilled,” she mentioned. “They’re simply very, very onerous to rent for.”
If you buy by means of hyperlinks in our articles, we may earn a small commission. This doesn’t have an effect on our editorial independence.
