At the first Reagan Economic Forum, Marc Andreessen, co-founder of a16z, shared his insights on AI and reindustrialization in the United States. He believes that the essence of AI is software, but its real revolution is not ChatGPT, but the upcoming era of robots. In the next decade, robots will become the largest new industry on the planet and drive a large number of mid-to-high-end employment opportunities. This article will delve into Andreessen’s perspective, analyzing how AI is moving from software to hardware and how this shift affects manufacturing, economic structure, and social policy in the United States.
At the first Reagan Economic Forum, which ended not long ago, a16z co-founder Marc Andreessen engaged in a systematic discussion around AI and the strategic inflection points of reindustrialization in the United States.
Andreessen said that we are at a profound turning point in technology, especially in AI. AI is no longer limited to models and cloud computing, but will also move towards “embodiment” – that is, robots.
In the future, there will be billions of robots operating, laboring, manufacturing, and serving in reality, and we need to vigorously develop future manufacturing jobs that will revolve around the design and manufacturing of robotic systems.
AI is moving from a bipolar confrontation between the United States and China to a hardware battle on the manufacturing side. China has effectively copied the path of the “American System”, rebuilding its industrial capacity through protectionism and industrial clustering.
Whether the United States wants to continue to be an industrial superpower is a policy choice, not a technical one. In the future, AI-driven economies will present a different structure from the previous service economy, which will affect policy arrangements in trade, immigration, regulation and other aspects.
The rise of AI and energy may usher in a new industrial era, but current U.S. policy is divided between manufacturing and technology acceleration. On the one hand, the traditional manufacturing revival faction tends to promote the return of heavy industry through tariffs and industrial incentives; on the other hand, there is a technological accelerationist faction, which focuses more on the new economy driven by chips, AI, and energy. Both paths are also reflected in the Trump camp.
B-end product manager’s ability model and learning improvement
The first challenge faced by B-end product managers is how to correctly analyze and diagnose business problems. This is also the most difficult part, product design knowledge is basically not helpful for this part of the work, if you want to do a good job in business analysis and diagnosis, you must have a solid …
View details >
The essence of the “American System” is not just tariffs, but the entire industrial ecosystem to achieve industrial clusters and growth, including a complete chain composed of OEMs and thousands of parts suppliers. The rise of British industry relied on heavy protection and light exports, and then moved towards free trade. The United States has also moved from high tariffs to “reciprocity” by McKinley, first by protecting and cultivating capabilities, and then using tariffs as a bargaining chip to obtain foreign markets.
From 1870 to 1970, the United States experienced two rounds of rapid growth. The first round was driven by the second industrial revolution, and the second round was driven by the space race and electronics and computer technology. But since the 1970s, U.S. economic growth has declined sharply, shifted to services and financialization, and actively deindustrialized, all of which are the result of a series of clear policies. This transformation, while leading to knowledge-intensive prosperity in Silicon Valley, Austin, and others, came at the expense of a broader economic foundation.
Low growth and deindustrialization have exacerbated populist sentiment in the United States, because in the absence of economic growth, the allocation of social resources becomes a zero-sum game, leading to social tearing. These problems are more evident in the cultural divide between the city and the countryside: the big cities have become a polar structure between the high-educated elite and the service bottom, while the middle class is squeezed out of the city and returned to the countryside, only to find that the countryside has no new economic opportunities.
The root of today’s urban ills is that the “knowledge economy” concentrates all growth in the city, creating extreme imbalances. Resources were dispersed in the agricultural or industrial age, but the knowledge economy concentrated elites in Silicon Valley, New York and other places, raising the cost of living in cities that middle-class families could not afford.
The final result is that only the elite and low-income classes remain in the city, and the middle class, which truly maintains social stability, is systematically excluded.
This is a geosocial deconstruction brought about by institutional reconstruction. Whether it is the “nationalization” of Microsoft and Amazon executives in Seattle by local politicians, or the systemic collapse of San Francisco, they are all direct products of “financialization + deindustrialization + extreme concentration of knowledge work”. The same is true in Europe, where the structure of “high-end elites + customer bottom” has also appeared in Paris, London and other places, and the entire West has entered a dangerous tipping point of “middle class flight and urban imbalance”.
These are choices, not irresistible technological forces or historical fate. We always hear that manufacturing is obsolete and unnecessary, but the truth is that the deindustrialization and energy transition that the United States has experienced over the past few decades have been the result of concrete policy choices.
For example, Nixon’s “Project Independence” in 1971, which originally planned to build 1,000 nuclear power plants by 2000, would fully switch the U.S. power grid to nuclear energy, thereby accelerating the popularization of electric vehicles and getting rid of energy dependence in the Middle East, but the plan failed due to his own EPA and NRC.
Now Washington is reevaluating a series of new policy options, including nuclear energy, rare earths, and urban construction. For example, in Solano County, California, where we have purchased land four times the size of Manhattan to build a new city, the window for these policy projects is reopening.
AI is a key variable and one of the strongest strategic leaders in the United States right now. But this requires policy coordination, otherwise it will be stifled by regulatory legislation in the states. For example, California almost legislated to ban AI half a year ago, which is a typical “self-destructing Great Wall”. Europe has largely sentenced AI to death through excessive regulation, and many AI talents are migrating to the United States as a result. AI is happening in the United States today, and the window of opportunity is fully present, and the key is whether we seize it or not, it is our choice, not fate.
AI is essentially software, and current applications, such as ChatGPT, are essentially mobile applications and have not yet created a wave of physical employment. But the next wave has already begun, and that is “embodied AI”, that is, robots. This trend is already being realized in the field of drones and autonomous driving: drones can fly autonomously, and autonomous vehicles are already in operation.
The next decade will usher in the large-scale landing of general-purpose robots such as humanoid robots, forming the largest new industry on the planet, which may be larger than any industry in human history, and requires a large number of mid-to-high-end manufacturing positions that are designed, manufactured, and deployed.
The correct path is not to “bring back the old manufacturing industry”, but to “build the future manufacturing industry with all its might” and build a super-automated factory and robot supply chain.
Elon Musk’s “alien dreadnought” style gigafactory will become the standard model, with thousands of new industrial categories and regional capacity deployments behind it. This will lead to the third or even fourth industrial revolution in the United States, unlocking a whole new level of economic growth and fundamentally responding to the question of “how to revitalize the rural economy”.
Only by moving from “software” to “hardware” can AI solve the tear between urban and rural areas and truly achieve a balance between broad employment and industry. Stopping at software alone will only make San Francisco, New York, Los Angeles, and Austin richer, and will not lead to new development paths for the countryside.
The new industrial revolution will not only rebuild the American manufacturing system, but also reconstruct the entire defense industry system. From fighter jets, missiles, radars to supply chain systems, AI-driven system reconstruction is required, and the industrial-level opportunities brought by the combination of AI + hardware have emerged. With the fate of large factories and unstable models, how can AI entrepreneurs find verifiable market opportunities?
On the issue of immigration, “low-skilled” and “high-skilled” immigrants need to be discussed completely separately. Some current policy logic is paradoxical and absurd, such as claiming that AI will replace a large number of jobs while calling for the introduction of another 100 million low-skilled immigrants from the third world. Under the wave of AI + automation, the marginal utility of low-skilled labor is decreasing, and what we need more is the talent density and R&D capabilities to build new industries.
What is really urgent is a change in the policy of “high-skilled immigration”. The talent base for AI, robotics, and large-scale systems engineering is extremely scarce, and the United States must attract the brightest minds in the world to build the future. The current series of systems around university expatriate students, visa policies, affirmative actions, etc., have invisibly had a great crowding out effect on middle-class families, especially those smart children from non-coastal, traditionally white communities, which are almost impossible to enter top universities, which will further exacerbate social divisions.
The admission system of top universities has evolved into a structure with a double tilt of “white DEI” (patron + sports recruitment) and “minority DEI”, in which the middle class is completely marginalized. Behind this is not only a change in the immigration structure, but also an institutional exclusivity driven by political correctness, which is profoundly changing the opportunity structure of American society.
The root cause of the problem is not geographical differences, but the structural closure and incompetence of core systems such as education, housing, and medical care for decades. However, the positive side is that there are still a large number of talents in the United States that have not been fully absorbed by the education system and industrial system for a long time. They have potential, but they have been systematically cut off from opportunities. The high productivity growth brought about by the AI and robotics revolution is expected to re-absorb these groups if the policy is designed correctly, forming a broad sense of “technology participatory growth”.
The real bottlenecks include energy (nuclear power restart), chips (production capacity and equipment), critical minerals, excessive regulatory systems, etc., which are clear and clear, and the key lies in implementation and policy consensus. Over the past few decades, the cost of industries deeply touched by technology – such as television, video games, and consumer electronics – has dropped sharply, while the three core pillars of the American dream – housing, health care, and education – have continued to soar in price, because all three are distorted by the triple mechanism of high regulation, government supply constraints, and demand-side subsidies.
He also said that the high-growth environment makes political problems easier to solve, makes the country more united and optimistic, and also makes society believe again that “the next generation can live better”. There are no cases of solving structural out-of-control problems such as medical care.
For example, eye surgery (such as laser correction) is one of the few non-government-led, non-medical insurance covered, and self-funded services in the medical field, and its price curve is similar to Moore’s Law, with rapid technology iteration, declining prices, and continuous improvement of service quality. This is a direct embodiment of technological access + deregulation, and it is also a paradigm sample of possible changes in healthcare, education, housing and other fields
This battle is not about “whether technology can do it”, but “whether society is willing”. The country must re-choose whether to promote real technology penetration and institutional restructuring to achieve a “technological re-industrialization” future with affordable housing, accessible education, and inclusive healthcare.
