You started the article with China's focus on the long term - 10 years.
This is the core of the United States problem.
The current AI bubble is an example. The talk is already around the near term return for the trillions in data center investment. Instead of doing an analysis of the model to determine how to optimize for less CPU usage, most AI LLMs are looking for near term return by just scaling the poor performing models. Deepseek was an example of tuning the LLM versus just investing in hardware. Classic for China.
Look also at what happened to Intel. They as a company were cornered by investors expectations of 60 point margins. TSMC had margins of 5 points. Intel could never focus on becoming a leading fab because the investors were pushing to maintain the 60 point margin. As a result, Intel is almost out of business.
The US innovation cycle is looking for 10x returns in a 3-5 year exit. That can not support developing high tech manufacturing that is 5 points margin.
The bottom line, in the US people want to get rich in 3-5 years through an exit and move on. In China, founders stay with the companies throughout their life. So do employees.
Unless the US moves to change the business model, we cannot compete in high tech manufacturing unless it's totally funded by the government similar to the defense industry. Trump suggested that corporations should only do corporate updates every 6 months and got blasted. How about moving to yearly updates with huge tax advantages for investments that have a 10 year return? Its a business conundrum.
The discussion of the issue is excellent but the prescribed solution makes no sense. Just as for rare earths and magnets (also critical for robots), suggesting that the US and allies should recreate supply chains already optimized in China appears wildly unrealistic. How did this approach work for 5G? Answer; not at all. A new approach and thinking is required here, such as how to partner with Chinese firms to contribute to supply chain development and how to pursue a realistic China policy that ensures long term access to the technologies of the future. The same old thinking will result in the same old outcomes...everything is not a race with China...
America risks losing the Robotics and AI race because we keep using a single-track strategy. We invest heavily in AI software, but not in the minerals, markets, and supply chains that make AI possible. China is doing all of it, expanding into fast-growing markets, financing exploration abroad, and securing copper, cobalt, lithium, and nickel. Meanwhile, the U.S. has put billions into AI, but very little into processing capacity or early-stage exploration. It takes only about $500k to explore for copper in Central and Southern Africa, and China is funding it.
If American technology companies don’t invest in the mineral supply chain of the future, we will have world-class AI, but no materials, no markets, and no resilience to sustain leadership.
Acceptance is important in life - we cannot always win everything , all the time ! In Sports too, you don't always win ! A loser is someone who thinks they cannot lose !
Make peace with History - For 2500 years of known Economic History ( Ref. Angus Maddison ) , India and China were #1 or #2 economies alternating for 2300 of those years , last time until 1850 . This 200 year period between 1850 - 2050 will be remembered as a Historical Anamoly, mostly due to Colonial Invasions by Europe of Asia, African and Americas. The world is re- balancing back to normal. Being #2 or #3 out of 200 countries , is still Not Bad !
America is at risk of losing the global race in Robotics and AI because we continue to approach these national priorities with a single-track strategy. Winning requires a multi-dimensional approach, one that includes expanding U.S. market dominance in regions with fast-growing populations, building resilient supply chains both at home and abroad, and securing the minerals that power advanced technologies.
Consider this: look at how much capital has flowed into AI models and software, versus how little investment has been made in processing capacity for copper, cobalt, lithium, and nickel, the minerals that AI, EVs, robotics, and energy storage all depend on. Look at how little U.S. capital is going into expanding new markets for AI adoption, especially in emerging economies. Meanwhile, China is doing all of this simultaneously.
It costs roughly $500,000 just to fund early-stage exploration for copper and other critical minerals in regions like Central and Southern Africa. China is financing it. China is securing future supply. China is building influence and new markets. Where are the American technology companies investing in the mineral supply chain of the future? Without parallel investments in raw materials, processing, new markets, and global capacity building, America risks building brilliant AI systems,only to find we lack the minerals, markets, and resilience needed to lead.
This article complains about labor and regulation, and ignores the 500 ton elephant in the room. Wall Street requiring financial profits that expand quarter after quarter.
Chinese companies invest more, and return lower profits to shareholders too.
The main point is: do the "West" still exists? Are US still believe in West? I don't think so, the main example is the trade war against what you call "Allies". Many countries and people starting to think that "West" is not the future
It's often said the West is 10 years or more ahead of China in the semiconductor industry. How many years ahead is China in drones and Robotics? Different Venture capitalists will tell you different things. The United States thinks it can compete with China in an infrastructure project like AI data centers? How many years behind is its energy grid? Clean energy? Capabilities to build in space? Saying we must win is not good enough.
"But our domestic capacity, when combined with those of allied states like Germany, Japan, and South Korea, can still make up almost the entirety of a robust robotics supply chain, and those weaknesses can be identified and addressed."
Sigh. You guys sound like CSIS. If you want to have impact, you have to accept that there is no interest from the USG in "combining" supply chains except when the leading firms of those countries can be coaxed into moving production to the US. I get that what you're proposing is actually the better solution, but key players prioritise aesthetic industrialisation to competition.
You argue that the regulatory environment > industrial clustering, but some of the reasons you cite have little to do with regulation per se and more with the ease of human capital spillovers in China. This has been seen broadly for a long time; people always want to be their own boss. New firms enter all the time in part because founders are not shackled by restrictive contracts. Innovating faster is seen as better than patent litigation. This is made possible by the ease of setting up companies, but without the underlying innovation culture it wouldn't work.
I am from Singapore, and for this article to say China leapfrogged Singapore and Korea tells me some LLM wrote this article hallucinating. Singapore is not anywhere close to the acme in this race. I am sure there are other inaacuracies in this hastily generated article.
You started the article with China's focus on the long term - 10 years.
This is the core of the United States problem.
The current AI bubble is an example. The talk is already around the near term return for the trillions in data center investment. Instead of doing an analysis of the model to determine how to optimize for less CPU usage, most AI LLMs are looking for near term return by just scaling the poor performing models. Deepseek was an example of tuning the LLM versus just investing in hardware. Classic for China.
Look also at what happened to Intel. They as a company were cornered by investors expectations of 60 point margins. TSMC had margins of 5 points. Intel could never focus on becoming a leading fab because the investors were pushing to maintain the 60 point margin. As a result, Intel is almost out of business.
The US innovation cycle is looking for 10x returns in a 3-5 year exit. That can not support developing high tech manufacturing that is 5 points margin.
The bottom line, in the US people want to get rich in 3-5 years through an exit and move on. In China, founders stay with the companies throughout their life. So do employees.
Unless the US moves to change the business model, we cannot compete in high tech manufacturing unless it's totally funded by the government similar to the defense industry. Trump suggested that corporations should only do corporate updates every 6 months and got blasted. How about moving to yearly updates with huge tax advantages for investments that have a 10 year return? Its a business conundrum.
The discussion of the issue is excellent but the prescribed solution makes no sense. Just as for rare earths and magnets (also critical for robots), suggesting that the US and allies should recreate supply chains already optimized in China appears wildly unrealistic. How did this approach work for 5G? Answer; not at all. A new approach and thinking is required here, such as how to partner with Chinese firms to contribute to supply chain development and how to pursue a realistic China policy that ensures long term access to the technologies of the future. The same old thinking will result in the same old outcomes...everything is not a race with China...
America risks losing the Robotics and AI race because we keep using a single-track strategy. We invest heavily in AI software, but not in the minerals, markets, and supply chains that make AI possible. China is doing all of it, expanding into fast-growing markets, financing exploration abroad, and securing copper, cobalt, lithium, and nickel. Meanwhile, the U.S. has put billions into AI, but very little into processing capacity or early-stage exploration. It takes only about $500k to explore for copper in Central and Southern Africa, and China is funding it.
If American technology companies don’t invest in the mineral supply chain of the future, we will have world-class AI, but no materials, no markets, and no resilience to sustain leadership.
It is one of the best, if not the best, in robotics matters. Thank you for clarifying and making us understand today's robotic situation.
Acceptance is important in life - we cannot always win everything , all the time ! In Sports too, you don't always win ! A loser is someone who thinks they cannot lose !
Make peace with History - For 2500 years of known Economic History ( Ref. Angus Maddison ) , India and China were #1 or #2 economies alternating for 2300 of those years , last time until 1850 . This 200 year period between 1850 - 2050 will be remembered as a Historical Anamoly, mostly due to Colonial Invasions by Europe of Asia, African and Americas. The world is re- balancing back to normal. Being #2 or #3 out of 200 countries , is still Not Bad !
Lawyers XOR builders?
Takers XOR Makers?
Hint; there’s no “legal” solution.
America is at risk of losing the global race in Robotics and AI because we continue to approach these national priorities with a single-track strategy. Winning requires a multi-dimensional approach, one that includes expanding U.S. market dominance in regions with fast-growing populations, building resilient supply chains both at home and abroad, and securing the minerals that power advanced technologies.
Consider this: look at how much capital has flowed into AI models and software, versus how little investment has been made in processing capacity for copper, cobalt, lithium, and nickel, the minerals that AI, EVs, robotics, and energy storage all depend on. Look at how little U.S. capital is going into expanding new markets for AI adoption, especially in emerging economies. Meanwhile, China is doing all of this simultaneously.
It costs roughly $500,000 just to fund early-stage exploration for copper and other critical minerals in regions like Central and Southern Africa. China is financing it. China is securing future supply. China is building influence and new markets. Where are the American technology companies investing in the mineral supply chain of the future? Without parallel investments in raw materials, processing, new markets, and global capacity building, America risks building brilliant AI systems,only to find we lack the minerals, markets, and resilience needed to lead.
I have jumped on the robotic wagon as well. My beta will be out soon for a hospital bot badly needed. Steel Fist Robotics.
This article complains about labor and regulation, and ignores the 500 ton elephant in the room. Wall Street requiring financial profits that expand quarter after quarter.
Chinese companies invest more, and return lower profits to shareholders too.
This is a great write up and right on point.
Exactly. This is the year of the robot. And the revolution is already unfolding all around us.
Even Jensen Huang recently said, “Everything that moves will be robotic.”
We wrote more about what's in store w/all the data here: https://riskhedge.substack.com/p/the-year-of-the-robot
Good stuff
The main point is: do the "West" still exists? Are US still believe in West? I don't think so, the main example is the trade war against what you call "Allies". Many countries and people starting to think that "West" is not the future
It's often said the West is 10 years or more ahead of China in the semiconductor industry. How many years ahead is China in drones and Robotics? Different Venture capitalists will tell you different things. The United States thinks it can compete with China in an infrastructure project like AI data centers? How many years behind is its energy grid? Clean energy? Capabilities to build in space? Saying we must win is not good enough.
++ Good Post, Also, start here Compilation of 100+ Most Asked System Design, ML System Design Case Studies and LLM System Design
https://open.substack.com/pub/naina0405/p/important-compilation-of-most-asked?r=14q3sp&utm_campaign=post&utm_medium=web&showWelcomeOnShare=false
"But our domestic capacity, when combined with those of allied states like Germany, Japan, and South Korea, can still make up almost the entirety of a robust robotics supply chain, and those weaknesses can be identified and addressed."
Sigh. You guys sound like CSIS. If you want to have impact, you have to accept that there is no interest from the USG in "combining" supply chains except when the leading firms of those countries can be coaxed into moving production to the US. I get that what you're proposing is actually the better solution, but key players prioritise aesthetic industrialisation to competition.
You argue that the regulatory environment > industrial clustering, but some of the reasons you cite have little to do with regulation per se and more with the ease of human capital spillovers in China. This has been seen broadly for a long time; people always want to be their own boss. New firms enter all the time in part because founders are not shackled by restrictive contracts. Innovating faster is seen as better than patent litigation. This is made possible by the ease of setting up companies, but without the underlying innovation culture it wouldn't work.
Exponential Views tracking of China’s solar growth is an important dimension of this challenge.
I am from Singapore, and for this article to say China leapfrogged Singapore and Korea tells me some LLM wrote this article hallucinating. Singapore is not anywhere close to the acme in this race. I am sure there are other inaacuracies in this hastily generated article.