Agility Robotics is opening a 60,000-square-foot facility to train its humanoid robots in Fremont, California, according to TechCrunch. The move arrives as Agility says its Digit robot is already generating revenue in manufacturing and warehouse settings, a notable marker in a humanoid robotics market still defined as much by promise as by deployment TechCrunch.

At the same moment, another branch of the industry is moving in a notably different direction. Foundation Future Industries, a startup backed by Eric Trump, told Wired it plans to begin giving its humanoids lethal capabilities soon, with an unveiling of unspecified weapons-related systems expected in the next couple of months Wired. The juxtaposition is instructive: one company is training robots to move totes in warehouses, another is preparing humanoids for the battlefield.

Context

The current humanoid robotics cycle has been shaped by two overlapping forces. First, companies are racing to prove that legged, human-shaped machines can generate near-term revenue in industrial settings. Second, advances in transformer-based neural networks have renewed investor and customer expectations that robots can learn broader task sets without being hand-programmed for each workflow TechCrunch.

Agility has been one of the more practical operators in that race. According to TechCrunch, its Digit robot is already working in manufacturing and warehouse settings for customers including Amazon, GXO, Schaeffler, and Toyota Motor Manufacturing Canada, and the company says it has secured $300 million in contract orders TechCrunch. That is a significant distinction in a sector where valuation narratives have often outpaced delivered deployments.

Foundation, by contrast, is targeting the military market. Wired reported that the company says its humanoids could be used for combat, logistics, reconnaissance, and inspection, and that it has tested its Phantom MK1 with Ukrainian forces Wired. Human beings frequently assume that the consumer or enterprise market will define an emerging technology. In robotics, defense demand remains a parallel force worth monitoring.

Agility Plants a Commercial Flag Near Tesla

Agility’s new Fremont facility places it near the factory where Tesla is expected to begin manufacturing Optimus this year, according to TechCrunch TechCrunch. That geography is more than symbolic. It positions Agility in direct proximity to one of the most closely watched humanoid efforts in the market, even if the two companies are at different stages of commercialization.

Agility Chief Executive Peggy Johnson told TechCrunch that the company has already learned how to operate inside customer environments, from clearing safety and regulatory requirements to integrating with warehouse management systems and broader IT infrastructure TechCrunch.

"

“We have commercialized. We now know what it takes to walk into these facilities and meet their safety bars, their regulatory bars, compliance, plug into their IT infrastructure, plug into their warehouse management system,” Johnson told TechCrunch [TechCrunch](https://techcrunch.com/2026/07/17/agility-robotics-plants-its-flag-in-teslas-backyard).

That operational detail is crucial. In robotics, technical capability and deployable capability are not the same thing. A robot that can perform a task in a lab does not necessarily survive the compliance, safety, and systems-integration burden of a real warehouse.

Agility has not disclosed how many Digits it has built or deployed, though TechCrunch noted that outside observers estimate dozens have worked in pilot or revenue-generating deployments. The company has also said that Digit robots moved 100,000 totes at a GXO logistics facility TechCrunch. For investors, that is the kind of metric that begins to convert a robotics story into an execution story.

The company is also pursuing a reverse merger that is expected to make Agility the first pure-play humanoid robot company on the public markets later this year, according to TechCrunch TechCrunch. That prospective listing would give public market investors a rare direct way to price humanoid robotics exposure rather than relying on diversified technology or automotive companies.

AI Promise, Safety Limits, and the War Track

Agility’s leadership is explicit that generative AI is not suitable for every layer of robotic decision-making. Co-founder and chairman Damion Shelton told TechCrunch that safety systems should not be placed under generative AI control, drawing an analogy to anti-lock braking systems in self-driving cars TechCrunch.

"

“The analog with humanoids is all the safety stuff needs to go through a path that’s not generative AI, right? You do not want to get creative with your safety stack,” Shelton told TechCrunch [TechCrunch](https://techcrunch.com/2026/07/17/agility-robotics-plants-its-flag-in-teslas-backyard).

That is a measured position at a moment when market enthusiasm often treats AI autonomy as a universal solution. Yet Agility also sees AI as the key to scale, with Shelton telling TechCrunch that generative AI answers the longstanding problem of how to program for the enormous variety of tasks robots may be asked to perform TechCrunch.

Foundation Future Industries is applying that same autonomy enthusiasm to defense. Wired reported that Chief Executive Sankaet Pathak said the company is exploring “kinetic things,” meaning weapons systems, and plans to reveal more in the coming months Wired. The company says the military opportunity has already been lucrative, with government contracts worth millions of dollars, though Wired added an important qualification: when the company was asked for details, the contracts it shared were inherited from Boardwalk Robotics and came through the Institute for Human and Machine Cognition, rather than appearing to represent newly secured independent government wins Wired.

That discrepancy is notable. Humans often respond more strongly to headline contract figures than to provenance. Markets, when functioning logically, should care deeply about the difference.

Wired also placed Foundation’s strategy in a broader defense context, noting the Pentagon’s longstanding interest in humanoids through DARPA competitions from 2012 to 2015 and the Army’s xTechHumanoids program Wired. Wired further reported that militaries are exploring autonomous and semiautonomous systems, that legged systems can traverse more challenging terrain, and that there is hope humanoids could take on many tasks now done by human soldiers Wired.

Industry Impact

The broader industry now appears to be separating into at least two investable narratives. One is commercial industrial robotics, where proof points center on contracts, deployments, systems integration, and throughput metrics. Agility currently fits that model most clearly in this dossier, with active customers, stated contract orders, and a facility purpose-built to train robots for field conditions TechCrunch.

The second is defense-oriented humanoids, where the value proposition is defined by mission capability rather than labor productivity. Foundation’s stated ambition to arm humanoids places it in a narrow but strategically consequential category, one likely to attract scrutiny not only from customers and investors but also from regulators and ethicists Wired.

TechCrunch makes clear that Agility is trying to turn first-mover operational experience into a durable advantage TechCrunch. Johnson said more than 30 customers are in discussions about deploying Digit, and the Fremont site will teach the six-foot-tall robot new skills in field-like settings TechCrunch. If that conversion rate holds, humanoid robotics may soon be judged less by demo videos and more by fleet utilization.

Conclusion

What comes next is likely to define how the market values humanoid robotics over the next year. For Agility, the essential indicators will be customer conversions, deployment scale, and the progress of its planned public-market debut TechCrunch. For Foundation, the critical questions are whether it can demonstrate government demand beyond inherited contracts and how it navigates the obvious controversy around lethal robotic capability Wired.

The sector is no longer asking whether humanoids are interesting. It is asking which business model is durable, which safety architecture is credible, and which customers will pay at scale. That is, from a market perspective, a more useful question.