On September 15, in Salem, Oregon, Agility Robotics pulled the sheet off Digit 5, the fifth generation of the humanoid that has spent almost three years doing unglamorous tote-hauling work on real production floors. The launch landed two weeks after Tesla’s Optimus ramp reporting dominated the news cycle, and the contrast between the two companies has rarely been this instructive. Tesla is betting that manufacturing scale and end-to-end AI will eventually absorb the deployment problem. Agility is betting the opposite: that the deployment problem, specifically the safety-certification and uptime layer, is the actual product, and that whoever solves it first owns the industrial market while the intelligence race sorts itself out.
The financial wrapper makes the wager concrete. Agility is going public through a merger with Churchill Capital Corp XI (NASDAQ: CCXI) at a $2.5 billion valuation, expected to generate more than $620 million in gross proceeds, including roughly $420 million from the SPAC trust and a $200 million PIPE led by Foxconn, according to The Robot Report. That same report noted the company disclosed $1.8 million in revenue ahead of the listing. A $2.5 billion valuation against $1.8 million of revenue is not a multiple you can justify with a spreadsheet. It is a claim about the future shape of the market, and Digit 5 is the evidence Agility is putting behind it.
What Digit 5 actually is
The hardware deltas matter more than they first appear. Digit 5 stands 5’11” (1.81 m), weighs 284 lb (129 kg), and reaches up to 7.2 feet (2.2 m), up from Digit 4’s 5.5-foot reach. That last number is quiet but commercially significant: it puts ground-level topline racking, the shelving universe designed around human height, within a humanoid’s workspace without ladders, step stools or facility retrofits. Agility’s sizing logic is that the robot should conform to the building, never the reverse.
Three upgrades define the generation:
Payload. A redesigned leg built around proprietary cycloidal actuator technology lets Digit 5 repeatedly lift 50 lb (22.7 kg), a 40% increase over its predecessor. CEO Peggy Johnson framed the threshold precisely: full coverage of single-person lift tasks in OSHA-regulated facilities. This is the regulatory framing doing marketing work, and it is smart. OSHA’s guidance treats repeated lifting above roughly 50 lb as an ergonomic hazard requiring controls. A robot rated exactly at that boundary does not just replace a worker’s task; it absorbs the compliance problem that comes with it.
Energy economics. A new battery system delivers 90 minutes of runtime and recharges in 9 minutes, a 10:1 run-to-charge ratio, up from Digit 4’s 2:1. The compounding effect is the headline: more than 20 hours of productive work per 24-hour day. Battery chemistry rarely gets discussed in humanoid coverage, but this ratio is arguably the most important unit-economics number in the entire launch. A robot that works 20 of 24 hours achieves an equipment utilization rate most capital assets in logistics never approach. Fixed cost per productive hour collapses as that ratio climbs, and it is why the 9-minute charge matters more than the 90-minute runtime.
Modularity. A swappable end-effector system with ISO-standard mounting flanges lets one chassis move between tote handling, machine tending and other work by changing grippers. The ISO flange detail is easy to skip and shouldn’t be: it means third-party tooling can bolt on without bespoke engineering, the same dynamic that made standardized tool interfaces a forcing function in industrial arms.
The cage-free thesis
The centerpiece of the launch is not lifting capacity or runtime. It is the claim in the announcement’s own words: Digit 5 is Agility’s first humanoid “engineered for cooperatively safe work at scale,” meaning close-proximity work alongside people without the physical barriers traditional automation requires.
The architecture has three layers. Safe human detection uses proprietary AI across multiple sensor modalities to continuously track people and respond by avoiding, stopping, or assuming a seated position, a low, stable posture that shrinks the machine’s footprint and fall energy. Safety cues, visual and auditory signals, communicate motion intent to nearby workers, treating the humans in the workspace as active participants in the safety loop rather than obstacles. And an independent safety controller, separate from the main compute stack, oversees person-detection events and triggers responses instantly.
Two facts give this architecture credibility that competitors’ announcements currently lack. First, Digit was the first humanoid to pass an independent field evaluation on a customer production line under OSHA-administered industrial safety standards, and Digit 5 builds from that baseline. Second, Agility is the first launch partner for NVIDIA’s Halos for Robotics platform, pairing its systems with NVIDIA IGX Thor compute and the Halos Core safety framework. Deepu Talla, NVIDIA’s vice president of robotics and edge AI, put the sequencing bluntly: safely operating around people and assets “is what moves humanoid robots from pilots to industrial deployments at scale.”
The strategic layer sits underneath the engineering. Agility is a project leader and active contributor to ANSI/A3 TR R15.108, the technical report for dynamically stable industrial mobile robots including humanoids in the US and Canada, and a contributor to ISO 25785-1, the first international safety standard for the humanoid category. The company is not merely preparing to comply with the rules of its industry. It is helping write them, and there is a long industrial history suggesting that the company holding the pen when a category’s safety standard is drafted converts that position into a durable moat: certification timelines, test protocols and evaluation precedents all tend to inherit the incumbent’s fingerprints.
The deployment ledger
Agility’s strongest evidence remains its operating history, which no competitor can currently replicate by announcement. Digit 4 has logged more than 65,000 hours across customer sites including GXO, Schaeffler, Amazon and Toyota Motor Manufacturing Canada. At GXO’s Flowery Branch facility near Atlanta, the fleet passed a cumulative 100,000-tote milestone at approximately 98% on-task accuracy.
It is worth pausing on what 98% “while on-task” does and does not mean, because precision matters more than optimism here. A 2% error rate on totes is a real operational number, the kind that determines whether a warehouse manager renews a contract. It says nothing directly about availability, mean time between incidents or recovery behavior, the metrics that Agility Arc, the company’s fleet management platform, tracks and exposes to customers as uptime, throughput and MTBI. The honest summary is that Agility has a certified, measured, renewing industrial deployment, which is more than anyone else, and it is still a tote-centric deployment expanding outward, not a general-purpose labor replacement.
That expansion is the explicit roadmap for Digit 5’s capabilities: depalletizing inbound goods, machine tending, kitting, sequencing, quality inspection, and palletizing outbound, with Arc coordinating handoffs to existing AMRs, conveyors and warehouse management systems. One robot class covering the material spine of a facility, orchestrated as a fleet rather than demoed as an individual.
The money and the mismatch
The order book is the launch’s commercial proof point: more than $300 million in multi-year customer orders as of May 2026, subject to contractual milestones, plus a stated pipeline across manufacturing, warehousing and logistics. Early access begins in the first half of 2027, with general availability by end of 2027, CE marking in progress for the first deployments outside North America in the EU and UK. Manufacturing runs through RoboFab, the 70,000-square-foot Salem facility purpose-built to assemble up to 10,000 Digit robots per year with more than 500 employees, supplemented by a new Fremont hardware and physical-AI development hub opened in July 2026.
Set against the SPAC filing, the picture sharpens into the central tension of the whole humanoid sector. Johnson’s quote in the announcement, “more than $300 million in multi-year orders and a growing pipeline,” and “value that compounds as Digit’s capabilities expand,” is a description of contracted future revenue contingent on milestones, layered on top of $1.8 million of actual recent revenue, offered to public-market investors at a $2.5 billion valuation. The bull case is coherent: safety-certified, cage-free operation plus 20-hour utilization plus a standards moat plus a 10,000-unit factory equals first-mover economics in a category where the incumbent automates the facility’s most persistent labor shortage. The bear case writes itself: milestone-contingent orders can slip, utilization guarantees do not survive contact with real incident rates, and SPAC proceeds have historically funded bridges, not destinations.
Both cases are stronger than the sector’s default narrative. The lesson of the past two years of humanoid hype is that the binding constraint was never leg design or actuator torque. It was the unglamorous stack of safety certification, fleet orchestration, charging logistics and contractual accountability, exactly the layer Agility has spent three years and 65,000 hours building while flashier programs chased keynote moments. Digit 5 is best read as the first humanoid deliberately engineered around the procurement checklist of an industrial buyer rather than the imagination of a retail one.
What to watch
Three verifiable signals over the next twelve months will tell you whether the wager is paying. First, the CCXI shareholder vote and PIPE close: money in the account converts a roadmap into a schedule. Second, early-access throughput data from first EU and UK deployments in the first half of 2027, which will test whether the safety architecture survives jurisdictions that regulate collaboratively operating machinery more strictly than OSHA does. Third, the ratio of Digit 5’s non-tote hours to tote hours by end of 2027, the truest measure of whether “full facility workflow” is a capability curve or a slogan.
The deeper industry question Digit 5 poses is uncomfortable for everyone racing on model capability. If cage-free certification and a 10:1 charge ratio get a humanoid onto a production floor years before a fundamentally smarter one clears safety review, then the market’s early winners will be decided by standards bodies and charging docks, not by foundation models. Intelligence can be patched in later, fleet by fleet, as Agility’s physical-AI training stack matures. Trust, once certified, compounds. That is the bet behind the $2.5 billion, and as of September 2026 it is the most defensible position in the field.
Sources: Agility Robotics announcement, The Robot Report on the CCXI SPAC and revenue disclosure, Forbes launch coverage, Humanoids Daily on Digit 5.