The U.S. may have the AI brains to power the next generation of robots—but building their bodies is proving much harder.

America wants to lead the next robotics revolution. Silicon Valley startups are developing increasingly capable models designed to give machines the ability to see, reason, manipulate objects and perform tasks in environments built for humans. Yet beneath that impressive software sits a stubborn industrial reality: many of the physical components required to build those robots are still cheaper, more readily available and more mature in China.

That gap could become one of the biggest obstacles facing the American robotics industry.

The problem is not that the United States lacks robotics talent. Quite the opposite. Companies such as Physical Intelligence are working on AI systems intended to operate across different classes of robots, from warehouse machines to humanoid platforms. The challenge is scaling those ideas beyond laboratories and prototypes into a manufacturing ecosystem capable of producing large numbers of reliable machines.

Physical Intelligence cofounder and researcher Sergey Levine recently highlighted the issue in a podcast discussion, arguing that access to dependable, inexpensive hardware is critical to accelerating U.S. robotics progress. A large amount of the hardware used in American robotics research currently comes from China because it is affordable, capable and already available at scale.

That creates an uncomfortable paradox for Washington.

The United States is trying to establish technological leadership in an industry that may shape the future of manufacturing, logistics, defense, healthcare and consumer services. But many of the components needed to make that future physically real are produced through supply chains in the very country Washington increasingly views as its strategic rival.

For years, discussions about artificial intelligence centered on algorithms, computing power and data.

Robotics changes the equation.

A software company can write an AI model and distribute it digitally around the world. A robot cannot exist only in the cloud. It needs motors, batteries, sensors, precision components, mechanical structures, control systems and other hardware.

Every one of those pieces has to be manufactured, tested, shipped and assembled.

That means America's robotics challenge is not simply about creating smarter machines. It is about rebuilding enough industrial capacity to manufacture them economically.

Industry executives have been unusually direct about the scale of the problem. Standard Bots CEO Evan Beard told a congressional hearing in April that producing hardware domestically can be dramatically more expensive than sourcing the same components from China. He said Chinese sourcing can be roughly five to ten times cheaper in some cases.

For startups operating with limited capital, such cost differences are enormous.

A robotics company can have an extraordinary AI model and still struggle to compete if every prototype costs substantially more to build than a comparable Chinese machine.

China's advantage is not limited to simply having large factories.

Its broader industrial ecosystem gives robotics companies access to supply chains that have been refined across multiple industries.

Electric vehicles are an important example.

Many technologies required for robots overlap with components used in EVs, including motors, batteries, magnets, power electronics and sophisticated manufacturing processes. China's enormous EV industry has therefore helped create infrastructure that can indirectly support robotics as well.

A June analysis cited by Business Insider found that China controls a significant share of capacity for robotics-related components, including about 90% of permanent-magnet processing. Those magnets are particularly important for high-performance motors used in advanced machines.

This is why the American robotics problem cannot be solved simply by opening more robot factories.

Factories need suppliers.

Suppliers need raw materials.

Manufacturers need skilled workers, tooling, energy and logistics.

In other words, robotics is an ecosystem.

The supply-chain problem would already be difficult without geopolitics.

But Washington increasingly treats advanced robotics and artificial intelligence as strategic technologies.

That has led policymakers to consider restrictions on foreign-built systems entering the American market.

In July, the Federal Communications Commission effectively blocked new foreign-made humanoids and other advanced robots from entering the U.S. market, citing national-security concerns. American manufacturers broadly welcomed the move, but some researchers and startups argued that restrictions can also make it harder to access the low-cost hardware that has helped U.S. robotics research progress.

That tension captures the dilemma perfectly.

Blocking foreign technology may protect domestic industry in the long run.

But doing so too quickly can increase costs for American companies today.

A startup that cannot obtain affordable motors, sensors or batteries may have to spend more money and time developing substitutes. Large companies may be able to absorb that expense. Young robotics businesses may not.

The emerging robotics race therefore has a fascinating division.

America appears particularly strong in the “brains”—AI models, software, research talent and advanced computing.

China has enormous strengths in the “body”—manufacturing scale, component production and industrial supply chains.

Winning the robotics era will likely require both.

A brilliantly intelligent robot that costs too much to manufacture will not transform the economy.

Likewise, a cheap robot without reliable intelligence may have limited usefulness outside highly controlled environments.

The real competitive advantage may go to the country that successfully combines the two.

That is what makes robotics different from previous technology races.

The winner may not be the country with the best single company. It may be the country with the strongest complete ecosystem.

The robotics challenge also forces the United States to confront a problem that extends well beyond robotics.

Over the past several decades, many forms of manufacturing moved abroad because overseas production offered lower labor and production costs.

For ordinary consumer electronics, that dependence can often be managed.

For strategic technologies, it can become a vulnerability.

If advanced robots eventually become essential to factories, warehouses, military operations and critical infrastructure, relying heavily on foreign-made components could create a strategic weakness.

That is why policymakers are increasingly interested in domestic production.

But rebuilding an industrial base is much harder than announcing incentives.

Companies must believe domestic factories will remain competitive for years.

Investors must accept the higher initial costs.

Workers need specialized skills.

And the supply chain must become large enough to produce parts consistently at competitive prices.

There is also a positive side to the problem.

If America can build a healthy robotics supply chain, the economic opportunity could extend far beyond humanoid robots.

A stronger domestic industry could support warehouse automation, advanced manufacturing, agricultural machinery, medical devices, defense systems and industrial automation.

It could also create demand for new generations of American companies specializing in motors, sensors, batteries, actuators, controls and robotics-specific materials.

In that sense, today's weakness could become tomorrow's industrial opportunity.

But that transformation will take time.

The public often sees robotics through spectacular demonstrations: humanoid machines walking, carrying objects or performing surprisingly complex tasks.

The less visible story is the one taking place inside factories.

Can a startup buy enough motors?

Can it source batteries reliably?

Can it obtain precision components without depending on a single foreign supplier?

Can those components be manufactured at a price that allows the final robot to compete?

Those questions may ultimately matter just as much as how intelligent the robot appears on a stage.

The United States has demonstrated that it can produce extraordinary AI research and software.

The next test is whether it can build the industrial machinery required to turn that intelligence into millions of physical machines.

America's robotics industry therefore faces a problem that sounds familiar from earlier technology battles: the country may have the ideas, the engineers and the capital, but critical parts of the manufacturing base remain somewhere else.

The robotics race is no longer simply a contest to build the smartest machine.

It is becoming a contest to build the entire supply chain behind it.

And unless America closes the gap between robotic intelligence and robotic manufacturing, the future of its machines could carry a very familiar label: Made in China.

Source basis: Yahoo Finance’s report based on Business Insider, plus current industry reporting and the cited McKinsey analysis.