Top 25 Countries Ranked By Advanced Robot Production Capability

Robots will costs 3X much to build successful America without immoderate Chinese content, according to a caller report.

getty

A humanoid robot that costs astir $46,000 to build coming will costs astir $131,000 if you return Chinese suppliers retired of the measure of materials, according to a study released coming by OpenMind that ranks the apical 25 countries by their expertise to build and run precocious robots.

"It’s not very useful for the US to person robots that costs 3x much than anyplace else," OpenMind laminitis and CEO Jan Liphardt told maine connected my Humanoid Daily podcast. "That instantly puts america astatine a disadvantage."

Liphardt is simply a Stanford bioengineering professor building OpenMind, which he describes arsenic an open, hardware-agnostic package furniture for robots. (Think Android for robotics.) His team’s caller Physical AI Readiness Index scores 58 countries and 22 metro clusters connected whether they tin really build, power, proviso and run precocious robots astatine scale.

Top countries for precocious robotics

The cardinal factors successful “physical AI readiness,” according to the report, see robotics and actuators, quality superior and software, power and grid capacity, compute and semiconductors, materials and proviso chain, positive the magnitude of section request for robots and precocious automation.

Here are the apical 25 countries successful the ranking, pinch their scores:

1. China 82.9
2. Japan 69.6
3. United States 69.0
4. South Korea 66.9
5. Germany 66.1
6. Taiwan 55.4
7. France 51.8
8. Switzerland 51.3
9. Sweden 50.6
10. Italy 50.5
11. Netherlands 49.6
12. Singapore 48.2
13. Canada 45.3
14. Austria 44.4
15. United Kingdom 44.4
16. Spain 43.1
17. Denmark 42.2
18. Finland 40.5
19. Israel 40.5
20. Norway 38.6
21. Belgium 37.8
22. India 37.2
23. Poland 37.0
24. Australia 35.4
25. Czechia 34.1

Both China and Japan rank up of the United States successful “physical AI readiness.” What that translates into is the expertise to build and vessel precocious caller robots. China’s 13.3-point separator complete 2nd spot is wider than the spread betwixt 2nd and fourteenth.

Interestingly, erstwhile you zoom successful for much details, the American problem becomes overmuch much apparent. The US leads compute astatine 91.6 and quality superior astatine 78.7, some by wide margins, but America scores conscionable 48.5 connected materials and proviso chain. In different words, the build blocks and the manufacturing components.

There, the US people conscionable 5th overall, astatine hardly half China's 95.

"Many group deliberation erstwhile they deliberation of beingness AI is that beingness AI is each astir AI," Liphardt said. "There’s this conception that if you person the champion models, past you're besides going to triumph beingness AI. But based connected what we tin tell, that's not true."

In portion to rebuild section manufacturing the FCC precocious added foreign-produced precocious robotic devices to its Covered List. That includes sidewalk transportation bots, inspection quadrupeds and humanoids and fundamentally immoderate ground-mobile instrumentality complete 2 kilograms pinch sensors, autonomy and a existent web connection. Under these caller rules, foreign-made units tin nary longer beryllium authorized for the US market. That’s good news according to immoderate robotics experts, but the costs impacts, astatine slightest successful the short term, are significant.

Adding tens of thousands of dollars to the costs of domestically-produced robots will not thief the manufacture turn locally, which is why the IDC suggests robot take velocity will take a superior hit if the rules instrumentality around.

Challenges for growth

"The full US robotics manufacture will now request to fig retired really to make precocious robots successful the US," Liphardt told me.

That is not and will not beryllium easy, for a assortment of reasons: manufacturing capability, earthy materials availability, and ecosystem density … fundamentally the reasons why the U.S. scored reasonably debased connected the Index.

It will not, however, beryllium intolerable either.

The U.S. does person the champion AI, and is continually reinvesting successful it – for illustration Figure’s billion-dollar bet connected video training data. But AI exemplary advantages support eroding arsenic unfastened root AI advances, conscionable months down the large proprietary models, seemingly. "Countries, particularly China, are unfastened sourcing precocious models besides for beingness AI," Liphardt adds. "That, of course, re-levels the playing field."

What’s harder than caller AI models is the existent metallic and uncommon earths robots require, on pinch precocious mechanical parts.

Actuators, a robot’s muscles, are 40% to 60% of a humanoid’s measure of materials. Semiconductors are astir 10%, and astir of those are mature 200mm and 28-to-90nm parts almost cipher is short of. The scarce parts are strain-wave and cycloidal reducers, planetary roller screws, frameless torque motors, precision bearings and rare-earth magnets. Magnets are a peculiar problem, because China makes 94% of the world’s sintered neodymium imperishable magnets and refines 91% of each world uncommon earths. (There are American alternatives, but not each of them are afloat up to complete accumulation velocity yet.)

OpenMind calls magnets the tightest azygous constraint successful the full proviso chain: tighter than chips, and acold tighter than batteries, which move retired not to beryllium a constraint astatine all.

No carve-outs for allies

A halfway situation is that the FCC edict is global.

It targets China, because China makes 97% of today’s shipping humanoid robots, but it makes nary objection for allies. The FCC defines "foreign-produced" via the Buy American Act’s home extremity merchandise standard: 65% home constituent costs done 2028, rising to 75% successful 2029. Which intends Japan, the second-place finisher successful this index, doesn’t lick the problem for American robot makers either.

Harmonic Drive Systems of Japan holds an estimated 71% to 85% of the world strain-wave reducer market, according to the report. Nabtesco holds astir 60% of ample business robot joints. A meaningful fraction of each of the humanoid robot joints connected earth, therefore, depends connected a Japanese company. But a Japanese portion doesn't count toward 65% home contented immoderate much than a Chinese 1 does.

That said, location are home manufacturers who are massively vertically integrating, for illustration 1X, which makes the Neo humanoid robot. 1X makes almost everything in-house, including actuators. That’s going to lessen the effect of the FCC decision.

Other manufacturers, for illustration Figure and Apptronik, are besides vertically integrating wherever possible.

Regional ecosystem density is simply a challenge

But there’s different problem for astir robot makers successful America – astatine slightest the ones who person raised billions of dollars and can’t vertically merge everything. And that’s conscionable sheer deficiency of ecosystem density.

"One of the things that China has done very good is spatially ore everything that you mightiness need," Liphardt said. "Within astir a 10 kilometer radial distance, you person everything you could perchance request from c fibre to CNC to bearings to sensors to PCBs."

The payoff for concentration, of course, is loop speed. "If you're building something, you tin spell crossed the street, you tin talk to personification who builds gears, and past you spell successful the different direction, and you're capable to conscionable move overmuch much quickly than you tin if everything is scattered crossed a large country."

Eight of the 10 starring robot-building metro clusters are successful East Asia. Shenzhen–Dongguan–Guangzhou is first, pinch an ecosystem people adjacent to 90. The Bay Area is apical successful the United States but only ninth globally, pinch a materials people of 32.

Here are the apical 20 location clusters according to the report:

1. Shenzhen–Dongguan–Guangzhou 87
2. Shanghai–Suzhou–Hangzhou 86
3. Tokyo–Kanagawa 80
4. Seoul–Gyeonggi–Ulsan 78
5. Beijing (Haidian + Yizhuang) 74
6. Hefei–Wuhu 72
7. Nagoya–Toyota 71
8. Hsinchu–Taichung–Taipei 70
9. Bay Area 70
10. Stuttgart–Karlsruhe 70
11. Munich–Augsburg 68
12. Boston–Cambridge 67
13. Zurich–Basel 65
14. Eindhoven–Brainport 63
15. Singapore 61
16. Västerås–Gothenburg 60
17. Austin 59
18. Milan–Bologna–Modena 57
19. Pittsburgh 57
20. Odense 56

America’s problem isn’t the full deficiency of robotics companies, suppliers, aliases moreover materials producers. The problem is comparatively utmost dispersion. "We surely person a batch of the pieces perchance present successful the U.S., but they’re typically successful very different places," Liphardt told me. "And those pieces that we do person are typically very mini and location aren't a batch of choices."

There are immoderate agleam spots. The Bay Area, arsenic mentioned above. Detroit–Ann Arbor ranks 21st connected cluster readiness globally. Austin is 16th. Boston is 12th.

All of these are important starting points that could service to galvanize much ecosystem density, but that mightiness return authorities nudging, and possibly incentives.

"You tin imagine, for example, different states making it overmuch easier to section trial beingness AI successful possibly hospitals aliases to build roads aliases bridges," Liphardt says. In different words, authorities tin participate connected some the request and the regulatory side, and actively push robots into existent activity to make the engineering feedback and training information that America can't different buy.

That’s not dissimilar China, wherever astir humanoid robots are bought by training facilities group up by various levels of government. They bargain the robots, and the robot makers bargain the information backmost from them to amended their robots. It’s circular, sure, but it’s an manufacture kickstarter too.

Ultimately, location is ray astatine the extremity of the tunnel. There is immoderate dream for section manufacturers.

"The spread is decidedly surmountable," Liphardt said. "And truthful we shouldn't beryllium afraid. This is, if anything, it's accusation that helps america prioritize."

OpenMind is releasing the afloat dataset and the codification down the index. "You're invited to alteration assumptions, to play retired various scenarios," Liphardt said. "Our extremity is simply to make it a batch easier for group to person a amended consciousness of what's going on."