Agility Robotics is trying to take humanoid robots public before the category has fully proven itself, and that is exactly why this SPAC deserves a closer look.
Agility Robotics is set to merge with Churchill Capital Corp XI in a deal valuing the company at about $2.5 billion, according to The Wall Street Journal. The combined company is expected to trade under the ticker AGLT, with more than $600 million in expected gross proceeds: $420 million from Churchill XI's trust account and more than $200 million from a PIPE led by Foxconn, the Taiwan-based electronics manufacturer that already backs Agility.
You should be skeptical of any SPAC by default. That structure carried too many companies onto the public market before they had enough revenue, enough discipline, or enough reality under the slide deck. Nikola became the warning label for the whole cycle, and Lucid showed how quickly an ambitious production story can turn into a public-market grind. Agility does not get a free pass because its robot walks on two legs.
But this deal is not just vapor wrapped in robotics language. Digit, Agility's humanoid robot, has been doing actual work. GXO Logistics put Digit into a Spanx facility in Flowery Branch, Georgia, where the robot moves totes from autonomous carts to a conveyor. Business Insider reported last year that Agility charges customers through a robot-as-a-service model that includes the robot, the work cell and the operating software. That matters more than a demo video. A robot standing on a stage is theater. A robot moving bins in a warehouse is at least a business starting to prove itself.
The Wall Street Journal reported that Agility's customers include Amazon, Schaeffler and Toyota Motor Manufacturing Canada, along with a logistics customer listed in its report as QXO. Older reporting and Agility's own commercial history point clearly to GXO's Spanx deployment as the company's most visible real-world proof point. Either way, the customer list is still short. It is also real, and in humanoid robotics that is not a small distinction.
Foxconn leading the PIPE is the most practical part of the deal. This is not a financial investor making a fashionable bet on physical AI. Foxconn builds hardware at enormous scale, and Reuters reported last year that Foxconn and Nvidia had been discussing the use of humanoid robots at a Houston AI server plant. Later reports tied Foxconn's Houston expansion to Nvidia server production and humanoid automation. If humanoid robots become useful on factory floors, Foxconn wants exposure from both sides: as a manufacturer and as a customer.
That is the kind of investor you want around a robotics company. Money helps. Manufacturing judgment helps more.
Agility's public-market problem will be scale. The Journal said the company is led by Peggy Johnson, the former Microsoft executive who also ran Magic Leap, and that its Salem, Oregon factory is designed to make as many as 10,000 units a year once fully running. That number sounds impressive, but capacity is not demand, and demand is not profitable deployment. The market will care how many Digits are actually working, how long they stay working, how much customers pay, and how often humans still have to step in.
Frankly, that is where humanoid robotics still has to earn the hype. Warehouses and factories already use plenty of automation, and a two-legged robot has to justify its complexity against cheaper machines built for narrower tasks. Digit's pitch is that it can work inside spaces designed for people, lift and move containers, and eventually take on more varied physical work. The first version of that story is credible. The full version is not proven yet.
Competition will not wait. Tesla keeps pushing Optimus. Figure AI has drawn major investor attention. Apptronik and Boston Dynamics are working the same broad frontier, while companies building robot foundation models are trying to make the software layer more capable. Agility's advantage is that Digit has moved from laboratory promise into commercial use earlier than most. That lead is valuable, but it is not permanent.
The public market is buying a race, not a finished company. Agility has a named robot, paying customers, an Oregon factory, Foxconn money and a valuation that is large without being absurd by late-stage AI standards. It also has undisclosed revenue, early deployments and a category famous for taking longer than investors want. If the SPAC closes, Agility will have to report its progress in quarters, not conference demos.
That is the useful thing about this deal. You can count robots. You can count totes. You can count customer sites. Once Agility is public, the story will stop being just about whether humanoid robots look ready and start being about whether they work often enough, cheaply enough and at enough sites to support a $2.5 billion company.
Also read: How to get into Y Combinator starts with the question most founders skip • Superhuman bets on trust as it buys AI detection startup GPTZero for its 19 million users and $30M ARR • How to Build a Sales Funnel for a Startup That Actually Converts
