Humanoid Robot Soldiers: How $24M Became an Arms Pitch

Founded in San Francisco in 2024, Foundation Future Industries is using roughly $24 million in research contracts from the Army, Navy, and Air Force to pursue a goal no other humanoid maker has attempted: fielding humanoid robot soldiers at scale. Earlier this year the company sent two unarmed Phantom units to Ukraine on logistics and scouting missions, a deployment it describes as the first time humanoids have operated in an active combat zone. Its CEO says armed versions could follow almost immediately if the Pentagon asks.

The escalation plan is not hypothetical. Sankaet Pathak, who previously ran a fintech company that ended in bankruptcy, has said publicly that he would build weaponized Phantoms on request, arguing that a combat robot needs to look intimidating and be the first unit through the door, deadly by design. The intimidation argument matters because it treats a robot's psychological effect on enemy troops as part of the product. Foundation has also signaled that it will unveil lethal capabilities for the platform in the coming months, a step that would move the company from defense research into direct weapons development. Pathak has separately argued that a fleet of humanoid robot soldiers could itself deter conflict by making aggression more costly.

The Phantom is a modest machine by industrial-robot standards. It measures roughly 5 feet 9 inches in height and can haul as much as 44 pounds of equipment, a payload that could include weapons, although no armed units have been built. In May the company released footage of the robot loading a mortar, and the two Phantom MK-1 units tested in Ukraine have been used to inspect and transport weapons, according to Foundation. The research contracts are framed around breaching, meaning the machines are meant to lead assaults on fortified enemy positions rather than stay in support roles. Foundation markets the platform as a way to take on that work while sparing American lives, and the mission is where the human form factor earns its keep in the company's telling, since urban and fortified terrain is built around human bodies.

From Research Contract to Production Roadmap

The production targets attached to that field test move the story beyond a routine defense pilot. Foundation plans to build 40 robots this year, about 10,000 next year, and tens of thousands by 2027, a ramp that would make it the largest humanoid manufacturer in the world if delivered. The company has said Phantom is meant to work alongside soldiers and factory workers alike, and it is pitching a leasing model to lower the upfront cost of adoption for both military and industrial customers.

The path to the Pentagon runs through unusual channels. Eric Trump is chief strategy advisor and an investor, giving the startup access to a presidential family at a moment when defense robotics is a stated national priority. Foundation has also acquired Boardwalk Robotics, a company with close ties to the Institute for Human and Machine Cognition, a Florida research center known for humanoid control systems, which brought in engineering talent and a research pedigree in one move.

The incentive structure explains the speed. A $24 million research portfolio is small by defense standards, but it converts into outsized marketing value when a startup can claim the first humanoid robot soldiers in a combat zone, presidential-family backing, and a path to US front lines within 18 months. The Ukraine deployment, whatever its operational results, is the proof point that justifies the next contract and the next funding round. Research contracts also fit the Pentagon's own incentives, letting the services explore a new category without committing to a full procurement program. That is the dynamic critics say rewards escalation over evidence.

The Trade-Offs of Arming a Humanoid

Those critics are organized. Democrats and human rights advocates have criticized the arrangement and warned about the dangers of autonomous weapons. Foundation's own framing sharpens the conflict: it presents the work as part of a race with China over military robotics, a positioning that turns restraint into strategic surrender and acceleration into patriotism.

The gap between what Phantom has demonstrated and what the roadmap promises is the central credibility problem for humanoid robot soldiers as a category. The Ukraine units handled logistics and scouting, work that smaller and cheaper systems already do with far longer field records. Lethal capability is an announcement rather than a fielded product, and no humanoid has yet demonstrated reliable autonomous combat in any theater. The drone battlespaces in Ukraine and Iran have already normalized autonomous systems in combat, shifting the question from whether machines fight to who controls them.

The trade-off extends beyond Foundation. Humanoid makers selling to factories and warehouses have spent years arguing that their machines are safe, predictable, and limited to structured environments. A high-profile push to weaponize the same form factor gives regulators and insurers a reason to treat the entire category with suspicion, and it hands lawmakers cover for stricter export and deployment rules that would also apply to commercial robots. The announcement also lands inside an international debate over autonomous weapons that has so far produced no binding rules. Foundation's bet on escalation therefore carries a cost that its competitors will pay as well.

Why Humanoid Robot Soldiers Draw Fire

On the evidence available, the company's production claims outrun its demonstrated capabilities. The company's own numbers make the gap visible: 40 units this year against a 10,000-unit claim for next year is a 250-fold jump. Reaching that figure would require manufacturing capacity and field-service operations that have not been shown, and the tens-of-thousands target for 2027 compounds that gap. What is real is the $24 million in contracts, the Ukraine deployment, and the political access; what is unproven is whether any of it translates into a working weapons program or a functioning factory.

None of this resolves the central question, which is what a Pentagon request for armed units would actually trigger. A formal request would force decisions the military has not made public, starting with how much autonomy is acceptable and who keeps the authority to fire. For investors, the distinction between a research contract and a procurement program is the entire ballgame: research money funds pilots, while only procurement funding validates the 10,000-unit targets.

The concrete milestones to watch are narrow:

1. Whether the Pentagon requests armed units, which would turn Pathak's stated readiness into an actual program.
2. The promised unveiling of lethal capabilities in the coming months, which would move the company's weapons work from planning into product.
3. Whether the 10,000-unit production figure survives contact with reality, since that number will determine if Foundation is a manufacturer or a story.

Why This Matters

For defense and robotics decision-makers, a $24 million research portfolio is small, but the precedent is not. If the Pentagon converts a politically connected startup's field test into a lethal-capabilities program, every maker of humanoid robot soldiers becomes a candidate for the same treatment and every regulator becomes more cautious. The 2027 production targets, met or not, will shape procurement policy, public trust, and the commercial market for machines that look like people. Watch the contract language more closely than the demo footage.

Related Articles

• Ukraine Bets on Combat Humanoid Robots Despite Mud and EW
• Hyundai's Atlas Humanoid Robots: The Unit Economics of a 25,000-Unit Factory Fleet
• Humanoid Becomes Europe's First Humanoid Robotics Unicorn with $152M Series A

✔Human Verified

Researched and cross-referenced against primary sources by the Bytevyte editorial team. This article was generated with the assistance of artificial intelligence and reviewed by the Bytevyte editorial team.