Agility Robotics has an opportunity to build a substantial business around repetitive industrial work. Its investment case rests on whether customers can deploy humanoid robots reliably, expand their use across facilities and justify the full cost of operating them. The introduction of Digit 5 makes that case more specific, with a new generation designed to handle heavier loads and work in closer proximity to people.

What interests me is the experience Agility has already gained in customer environments. A robot used in a warehouse must cope with the routine problems of production: misplaced containers, changing schedules, maintenance and interruptions. Learning to manage those conditions can be commercially valuable if it leads to better products and repeat orders.

The opportunity remains early. Agility needs to turn customer commitments into installations, revenue and eventually attractive returns on capital. My bullish thesis depends on that progression, rather than on the size of the humanoid market alone.

Understand the security before assessing the company

Agility announced a proposed merger with Churchill Capital Corp XI, which trades as CCXI, on June 24. The combined company is expected to use AGLT after closing. As of this article's research date, the transaction remains proposed; CCXI and AGLT should not be described as two independently listed Agility businesses.

The announced transaction values Agility at $2.5 billion before the new capital and anticipates more than $620 million in gross proceeds, including approximately $200 million of PIPE financing. The proceeds are subject to transaction conditions, and shareholder redemptions can reduce the cash received. The headline valuation is a transaction term, not a current stock-market valuation.

For investors, this means evaluating both the operating company and the deal structure. Completion, available cash, transaction expenses and the eventual share count all affect the investment outcome.

Digit 5 addresses practical deployment constraints

On September 15, Agility unveiled Digit 5 with a stated lifting capacity of 50 pounds, a battery designed for 90 minutes of runtime and nine-minute charging, and interchangeable end-effectors. Agility says the charging design can support more than 20 productive hours per day. These are company specifications and intended capabilities, rather than independently established performance across every customer site.

The improvements address identifiable commercial limitations. Greater payload can increase the number of suitable tasks. Faster charging can improve utilization. Changing the tool at the end of an arm can allow the same platform to perform different jobs, provided the corresponding software and workflow have been validated.

That last condition is important. Moving a standardized tote does not establish that a robot can reliably perform every form of machine tending or inspection. Each application has its own tolerances, failure modes and customer acceptance requirements. Investors should look for evidence that the additional capabilities work consistently in production.

The NVIDIA relationship has a defined technical purpose

NVIDIA's June announcement identifies Agility as its first partner incorporating elements of Halos for Robotics into a proprietary safety system. Halos combines industrial computing, sensor connectivity, software and a process intended to support inspection and third-party certification.

The commercial relevance is straightforward. A robot that requires extensive separation from employees can be difficult to introduce into an existing operation. Better detection and controlled responses may increase the number of suitable locations and reduce deployment complexity. However, the complete installation still needs an appropriate safety assessment; using NVIDIA technology does not automatically certify every application.

I view the collaboration as useful engineering support for Agility's product strategy. Its value to shareholders will depend on whether it helps deliver systems that customers can operate safely and economically.

Customer agreements provide evidence of demand

GXO announced a multiyear Robots-as-a-Service agreement with Agility in June 2024 after a pilot. Its release described Digit moving totes from other automated equipment onto conveyors at a SPANX facility, with Agility Arc coordinating the work. This is a defined commercial use case within an existing logistics operation.

In February 2026, Toyota Motor Manufacturing Canada signed a commercial agreement after its own pilot. The announced plan covered manufacturing, supply-chain and logistics work, with the companies continuing to evaluate further applications. That progression from evaluation to an agreement is a useful commercial milestone.

Neither relationship establishes adoption across every facility operated by those groups. The next evidence I would prioritize is expansion at existing customers, followed by replication at additional sites. A repeat deployment suggests that the customer has seen enough value to commit again.

Read the order figure carefully

Agility's June investor presentation describes more than $300 million of multiyear Digit 5 orders as of May 2026. Its footnote connects the figure to 1,000 robots under a three-year RaaS contract, subject to contractual milestones, with purchaser warrants vesting as robots are deployed. It explicitly distinguishes this value from current-period revenue.

That is a significant expression of demand, but it is not cash already earned or evidence of many independent customers placing equally large orders. The contract structure also makes delivery milestones and potential dilution relevant to the commercial assessment.

The bullish development would be timely conversion into operating robots and recognized revenue, followed by additional commitments from a wider customer base. Until then, headline order value should be assessed alongside the conditions required to realize it.

Manufacturing and service determine the economics

Digit is assembled at RoboFab in Salem, Oregon. Agility says the facility is designed for capacity of up to 10,000 robots annually at full scale. Designed capacity is not current output, and it does not establish that customers will absorb that volume.

A credible production ramp requires qualified components, repeatable assembly, testing and a service organization capable of supporting the installed base. As volumes increase, manufacturing efficiencies may reduce costs. At the same time, warranty claims and field failures can become expensive quickly.

Agility's presentation outlines both RaaS and customer ownership models, with software and maintenance services. The commercial appeal of RaaS is a lower initial commitment for the customer and revenue over time for Agility. The financial challenge is funding hardware and support before recovering the full investment.

Agility Arc supports deployment and fleet management, including workflow setup and troubleshooting, as described in GXO's announcement. Software can strengthen customer relationships when it becomes useful to daily operations. Its economic contribution still needs to be assessed together with the cost of keeping the robots productive.

Useful work must earn an acceptable return

The metric I would focus on is the total cost of reliably completing a task. That includes financing, maintenance, integration, charging and any human assistance required. A robot that can technically perform a task may still be a poor purchase if simpler automation does it more cheaply.

Humanoids become more attractive where existing layouts and varied tasks make specialized systems difficult to justify. The potential advantage is flexibility across work designed for people. Agility must demonstrate that this flexibility compensates for the complexity of a walking machine.

The financial starting point is demanding. MarketWatch reported, citing a regulatory filing, that Agility recorded a $138 million net loss in 2025 and disclosed substantial doubt about its ability to continue as a going concern without the planned transaction. This makes financing central to the thesis.

The milestones that matter next

Agility expects early access to Digit 5 in the first half of 2027 and general availability by the end of that year. I would track customer acceptance, productive utilization and the amount of assistance required once systems reach operating sites. Those measures will say more about commercial readiness than a demonstration alone.

I would also watch whether contract milestones are met, whether repeat customers expand their fleets and whether service costs improve as the installed base grows. A successful merger provides funding; it does not by itself solve manufacturing, reliability or customer-return requirements.

Competition will remain intense, including from conventional automation. A large opportunity can support multiple suppliers, but it can also attract aggressive pricing and heavy spending. Shareholders need value created per share, after the costs of growth and dilution.

Why Agility remains worth following

Agility offers a specific way to evaluate the industrial humanoid opportunity: observe customers adopting a robot for defined work, then assess whether deployments expand profitably. Its customer agreements and new product generation give investors concrete developments to follow.

The strongest bullish outcome is a business in which successful installations generate repeat orders, software and service revenue, while manufacturing improvements lower the cost of deployment. That could create a much larger company over time. Delivering that outcome requires considerable capital and execution, and the purchase price of the shares remains essential to any investment decision.