At the recent first Reagan Economic Forum, Marc Andreessen stated that the United States must take a leading role in robotics technology in the age of artificial intelligence, or else face the impact of a flood of Chinese robots. He noted that since the mid-20th century, the share of manufacturing in the U.S. economy has declined significantly. Andreessen called for building "Star Destroyer" factories to achieve deindustrialization and drive future economic growth.
Marc Andreessen is a renowned American entrepreneur, venture capitalist, and software engineer, hailed as one of the pioneers of the development of the Internet. With his keen insight into technological trends and forward-thinking thinking, he has become a thought leader in the field of tech investment. He actively advocates for software-driven innovation, famously stating that "Software is Eating the World," emphasizing the transformation of all industries by technology.
His speech at the Reagan Economic Forum this time exemplifies these characteristics. The Mind Observation Institute has compiled the speech content for reference.
Marc Andreessen: Hello everyone! Welcome to our first Reagan Economic Forum. It's an honor to be here with you. We often discuss the economic pillars and policies of President Reagan here—deregulation, tax cuts, and spending reduction, which remain crucial today. But today, I want to start from a longer historical perspective, talking about how the United States should respond to the current strategic turning point, and how artificial intelligence (AI) and deindustrialization will shape our future.
I have been studying the history of the U.S. economy recently, especially the "American System" of the 19th century. This system was first proposed by Alexander Hamilton when he served as Secretary of the Treasury. At that time, the United States had just experienced the Revolutionary War, with an economy primarily based on agriculture, while Britain, as the world's first industrial power, had overwhelming economic, technological, and military advantages. Hamilton had intense debates with Thomas Jefferson: Jefferson wanted the United States to maintain an agricultural and aristocratic rural economy, while Hamilton advocated for building an industrialized, urbanized, and financialized superpower. In the end, Hamilton's ideas prevailed, laying the foundation for the American System and pushing the United States to gradually become an industrial power in the 19th century.
By the late 19th century, especially during the presidency of William McKinley (1897-1901), the United States reached the climax of the Second Industrial Revolution. From 1870 to 1930, almost all the physical infrastructure we now consider modern was born during this period: airplanes, modern manufacturing techniques, mass-produced glass products, air conditioning, television, radio, automobiles... Look around, almost every piece of physical infrastructure comes from that era. McKinley's leadership made the United States a global industrial superpower, laying the foundation for its key role in both World Wars, especially as the "arsenal of democracy" in World War II.
The core of the American System is protectionism. Its logic is not merely relying on a single technology or company, but building a complete industrial ecosystem—there are not only three major automobile companies, but also thousands of parts suppliers. Britain also adopted similar protectionist strategies during the First Industrial Revolution, later shifting to free trade to promote exports. McKinley inherited this tradition, initially implementing high tariffs to protect American industry, but as American industry grew and export demand increased, he shifted to advocating "reciprocal trade," using America's high tariffs to force other countries to lower their tariffs and open up global markets. This model was later emulated by countries such as Germany, Japan, and South Korea, and China is currently running a similar "American System" script.
This raises a fundamental question: Do we want the United States to continue as an industrial superpower today? In the 19th century, this goal was clear, but now we are ambiguous about it. Current policy discussions—tariffs, trade, industrial incentives—are remarkably similar to those of the McKinley era, but we need to clearly decide whether to rebuild industrial advantages through protectionism or other tools.
Let's look at the trajectory of the U.S. economy. From 1870 to 1920, the U.S. economy was driven by industrialization, growing at three times the rate we see today, despite a smaller population and fewer technological tools. This was an environment of high immigration and high tariffs. From 1920 to 1970, the growth rate slowed slightly, about twice as fast as today, accompanied by the space race, the rise of electronic technology, and the computer industry. But from 1971, the year I was born, the U.S. economy saw a significant decline in growth, entering a low-growth phase. This is not only reflected in GDP, but also in productivity growth—the speed of technological change—slowing down.
The famous Ford motor production line in 1920
This low growth is closely related to our active choice of deindustrialization. Through a series of clear policies, we banned many industrial activities, shifting to a service economy and financialization. Knowledge economy centers like Silicon Valley and Austin have flourished, becoming hubs for global technology exports, and the AI revolution is also booming in the United States. However, this has not benefited the entire country, especially rural areas. Deindustrialization has led to slower economic growth, and low growth has fostered a zero-sum mindset—people feel that opportunities are getting fewer and resources can only be obtained through competition. This is the root of the rise of populist movements on both the left and right.
More dangerous is that economic transformation has exacerbated the divide between cities and rural areas. I was born in a rural area of Wisconsin, which was once an agricultural and light industrial region. Later, I moved to California, where I felt the strong prejudice of coastal elites toward the Midwest countryside. When I was in Wisconsin, I thought the "smart people" on the coast were much better than us; but when I went to California, I found they not only disliked us, but even looked down on us. This urban-rural division is not only economic, but also cultural and social identity, forming a major隐患 in American society today.
Our cities are becoming dysfunctional. Take New York, for example, where a candidate ran on a platform of "city-owned grocery stores," essentially reviving the "bread line" of the 21st century. Seattle is even more absurd: a city council member who advocates nationalizing Microsoft and Amazon was elected in the district where the executives live. In the San Francisco Bay Area, I live in a "epicenter" of dysfunction and collapse, with countless stories of governance failure. French geographer Christophe Guilluy analyzed this phenomenon in his book "The Twilight of the Elites": during the process of deindustrialization and financialization, agricultural or manufacturing economies require resources distributed nationwide, because natural resources, energy, and infrastructure are crucial. But in the knowledge economy, economic growth is concentrated in cities, like Silicon Valley, which attracts a large number of software developers, driving up housing prices—$3 million for a house.
The result is that cities are left with two types of people: high-income knowledge elites who can afford the sky-high houses; and low-income service workers dependent on public housing (such as Section 8). The middle class is completely pushed out of the city. For example, in the Bay Area, police officers or firefighters need to commute three hours a day to afford a house and family. There are more dogs than children in the city, because ordinary people can't afford a house, raise children, or even buy a car. This economic structure leads to local democracy in cities being dominated by progressive elites and the "client group" (such as immigrants or minorities), forming a "top + bottom" political alliance, while the middle class is pushed to the countryside. The problem is that the countryside lacks economic opportunities due to deindustrialization, leaving the middle class nowhere to go.
All of these are the results of policy choices, not irresistible economic or technological forces. In 1971, President Nixon proposed the "Independence Plan," aiming to build 1,000 nuclear power plants by 2000, transform the power system, accelerate the development of electric vehicles, and reduce dependence on Middle Eastern energy. This could have changed history, but the same year, the Environmental Protection Agency and the Nuclear Regulatory Commission he created hindered this plan, resulting in almost no new nuclear power plants being built. This is a consequence of a policy choice.
The good news is that we are now facing a major technological turning point—the rise of artificial intelligence. The United States leads in AI, but this is not inevitable. Europe, due to excessive regulation, has almost banned AI, with talent flowing to the United States; while China is our only real competitor. Currently, AI mainly exists in software form, such as mobile applications like ChatGPT, and has not yet created many middle-class jobs. But the next stage of AI—embodied AI, that is, robotics technology—will completely change everything. Drones have already shifted from manual operation to autonomous flight; self-driving cars, like Waymo in San Francisco, have become a reality—despite the cyberpunk-like future where you'll see self-driving cars passing people who have overdosed. Elon Musk's Optimus robot project shows that within the next decade, the robotics industry may become the largest industry on Earth, involving billions to hundreds of billions of various types of robots.
We should not try to revive traditional manufacturing, such as having workers stand on assembly lines tightening screws. What we need is future manufacturing—designing and producing AI-driven hardware, such as drones, self-driving cars, and robots. Musk's "Star Destroyer" factories—highly automated production facilities—will drive national economic growth, creating tens of millions to hundreds of millions of jobs, especially in rural areas. This will enable the United States to lead the third or fourth industrial revolution, and from a national security perspective, avoid being dominated by a robot world led by China. If we don't do this, the consequences will be profound.
Many people worry that AI will destroy job opportunities, but this concern is wrong. Over the past 50 years, we have not been in an era of rapid productivity growth or technological change, but rather a low-growth era. If AI develops as we expect or worry, it will greatly increase productivity and accelerate economic growth. If AI remains at the software level, cities will become wealthier, but rural issues will still remain unsolved. To solve rural problems, we need to convert AI into hardware, creating new industries through deindustrialization—computerized, AI-driven various hardware. This is not only the direction in which I and Joe (Lonsdale) are investing, but also the direction of rebuilding the defense industry base, such as building cheaper roads, tunnels, and infrastructure with AI and robotics technology.
Immigration policy is another complex issue. For the past few decades, we have pursued large-scale immigration. The tech industry has always advocated for attracting more high-skilled talents—many of my smartest friends are immigrants who have created great achievements. But large-scale low-skilled immigration has some negative impacts on the working class, especially in the context where AI and robotics technologies may reduce the demand for low-skilled labor. Historical data shows that the United States achieved high growth after limiting immigration from 1925 to 1970, indicating that high immigration is not a necessary condition for economic growth.
The debate over high-skilled immigration is heating up. We need talent to build AI and robotics, but current university admission policies and DEI (diversity, equity, and inclusion) measures distort this process. The proportion of foreign students in top universities has increased from 2-4% 50 years ago to 30-70%, taking away opportunities from local students. More concerning is that DEI policies lead universities and companies to recruit African immigrants (such as Nigerians) to meet diversity goals, rather than native African Americans. This not only affects white and Jewish groups, but also deprives students from the Midwest and South of opportunities. This policy exacerbates populist anger, with rural residents feeling "hated" by coastal elites. We have a large amount of underutilized local talent, especially in the Midwest and South. By reforming education and employment policies, we can tap into this potential and reduce over-reliance on high-skilled immigrants.
The bottleneck for achieving this vision is obvious: energy, critical minerals, and regulation. Energy is the foundation of AI and deindustrialization. We need nuclear energy and other clean energy sources. Rare earth metals and other resources are essential for hardware manufacturing. And regulation is the biggest obstacle—housing, healthcare, and education are controlled by powerful lobbying groups, hindering competition and technological application. For example, laser eye surgery (LASIK) has seen prices continuously decrease and quality improve due to marketization and technological advancement, proving that technology can break the trend of rising costs. But housing, healthcare, and education have seen price surges due to government restrictions on supply (such as zoning laws, university accreditation) and subsidies for demand (such as home purchase subsidies). Voters must realize that $3 million houses, $1 million college degrees, and medical costs accounting for 20% of GDP and possibly rising to 50% are unacceptable.
The good news is that these issues have solutions. Ezra Klein's new book "Abundance" reminds us that liberals also supported construction in the past—rockets during the Soviet era, skyscrapers, and cars were filled with optimism for the future. Now, some Democratic leaders are also promoting the concept of "building the future," opening up possibilities for bipartisan cooperation. Under the impetus of a high-growth environment and rapid technological development, these issues will be easier to resolve. Economic growth will bring optimism, reduce the zero-sum mentality, and make people believe that the next generation will live a better life—just as President Reagan said, "the morning of America."
We have also purchased land equivalent to four times the size of Manhattan in Solano County, California, planning to build a new city to provide a new model for economic growth and housing. These are choices we can make. As long as we are determined, the United States can lead the AI and robotics revolution, rebuild the industrial base, unleash the potential of local talent, bridge the urban-rural divide, and create a more prosperous and united future.
Thank you, everyone!
This article is exclusive to Guancha.cn. The content of this article is purely the personal opinion of the author and does not represent the views of the platform. Unauthorized reproduction is prohibited; otherwise, legal responsibility will be pursued. Follow Guancha.cn WeChat official account guanchacn to read interesting articles every day.
Original article: https://www.toutiao.com/article/7525991572521959976/
Statement: This article represents the personal views of the author. Please express your attitude by clicking on the 【Up/Down】 button below.