In this article
Sankaet Pathak and his startup Foundation Future Industries aim to build an American robot supersoldier rather than a domestic helper.
Lethal capabilities coming soon
Pathak, the CEO, plans to equip his humanoids with weapons systems within the next few months, though he declined to offer specific details. “We have some kinetic things we’re exploring,” he told WIRED. Beyond fighting, the company notes these units could handle logistics, reconnaissance, and inspection duties.
A long history of military interest
The US military has pursued humanoid technology for years. The Defense Advanced Research Projects Agency funded major contests between 2012 and 2015, while the Army runs a program called xTechHumanoids to fund relevant technology. Militaries worldwide are adopting autonomous systems, ranging from drones to compact vehicles. Legged systems can move through difficult terrain, and the hope is that robots could take on many tasks currently done by human soldiers. The war in Ukraine has served as a testing ground for these systems; Foundation says it has tested its humanoid, Phantom MK1, with Ukrainian forces.
Backed by Eric Trump
The company targets the military market and has secured government contracts worth millions. High-profile backers include Eric Trump, the president’s son, who serves as an investor and chief strategy adviser. Pathak claims Trump is an engineer at heart who does milling and other work at home.
“When you go up and you interact with these robots, and they fist-bump you, they high-five you, they follow your commands,” Eric Trump said during an interview with Fox Business on April 23. “You bring in AI autonomy, it’s going to change industry, going to change military application, it’s going to change hospitality. The uses are unlimited, and I think it’s a very beautiful thing.”
Contract details remain fuzzy
Foundation was founded in 2024. A few months later, it acquired Boardwalk Robotics, which worked closely with the Institute for Human and Machine Cognition (IHMC). During the Fox segment, the host mentioned a “$24 million contract with the Pentagon,” but the specifics are unclear. When asked for more information, Foundation shared details of two contracts inherited from Boardwalk and three that came through IHMC. The company does not appear to have independently secured new funding from the government.
Why the military wants robots
Some experts see promise in the niche. “If you put a military hat on, it makes a lot of sense, because it’s where soldiers still die—that first entry through a door,” says one roboticist familiar with Foundation who asked to remain anonymous. “If you look at Fallujah, the first Gulf War, you had several thousand insurgents hiding in 10,000 buildings and [US troops] just going door to door.”
“I think it is so close to feasible that I’m surprised they’re not already fielded,” they add.
The limits of autonomy
Like other humanoid companies, Foundation often portrays its robots performing tasks autonomously, but experts say fully autonomous robot soldiers are a distant dream at best.
“Right now, it is difficult to disentangle the current state of the art from the potential of the state of the art” with humanoids, says Robert Griffin, a senior research scientist at IHMC who led a project involving Boardwalk. “There’s a bunch of challenges, spanning the whole gamut of robotics, for the idea of building an actual human soldier,” Griffin says.
Humanoids have advanced thanks to cheaper motors, sensors, and AI algorithms that train systems to perform dynamic moves like parkour. But perception and navigation remain key problems when humanoids are placed in unfamiliar situations. Although they can demonstrate remarkable balance, they often need specialized training for different terrain. Physical manipulation, which is crucial for tasks like picking up a gun, remains a major unsolved challenge.
Rodney Brooks, a robotics pioneer and professor emeritus at MIT, says he expects it to take more than a decade for humanoids to operate reliably in complex and unfamiliar settings. Even in the lab, a combat humanoid would need to traverse a wide range of terrains and building types, navigate rubble on stairs, and get through blocked doors. “Going from a solid lab demo to initial deployment in robotics is always at least 10 years,” Brooks says.
Ethical questions and doomsday fears
Deploying new military autonomy raises ethical questions around reliability and a lack of human input in decisions about the use of deadly force. The notion of placing humanoid robots in combat conjures sci-fi images of The Terminator.
Pathak is dismissive. “From my perspective, the whole doomsday scenarios are very, very overblown for humanoids,” he says.
He also does not seem worried about efficacy and believes robots and other tech could make war more precise and efficient. “I wish I could end wars, but I don’t think I can,” he says. “But I can definitely contribute in making war more precise, reducing collateral damage as much as possible.”
That goal might be far off. Pathak says the next version of the company’s humanoid, Phantom MK2, will be the first version of the robot to be both waterproof and dustproof.




