At this stage, the winners of AI are all users, and the losers are companies with proprietary models.

Author: MD

Produced by: Mingliang Company

Recently, the well-known American podcast Invest Like the Best interviewed Marc Andreessen, co-founder of Andreessen Horowitz. In the interview, Marc and host Patrick delved into the significant transformations AI is reshaping in technology and geopolitics, discussed the open-source artificial intelligence of DeepSeek and its implications in the great power technology competition, and shared their views on the evolution of the global power structure and the overall transformation of the venture capital industry.

"Mingliang Company" utilized AI tools to promptly organize the core content of the interview. For the full text, please refer to the "Original Link" at the end.

The following is the interview content (edited):

Discussing DeepSeek, AI Winners and Losers

Patrick: Marc, I think we have to start with the most fundamental question. Can you talk about your thoughts on DeepSeek's R1?

Marc: There are many dimensions to this. (I believe) the United States is still recognized as the scientific and technological leader in the field of artificial intelligence. Most of the ideas in DeepSeek stem from work done in the past 20 years, and surprisingly, even 80 years ago in the U.S. or Europe. The initial research on neural networks began as early as the 1940s in research universities in the U.S. and Europe.

So, from the perspective of knowledge development, the U.S. is still far ahead.

But DeepSeek has done an outstanding job of applying this knowledge. They have also done something remarkable, which is to provide it to the world in an open-source format. This is actually quite astonishing because there has been a reversal of this phenomenon. You have American companies like OpenAI that are essentially completely closed.

Elon Musk's part of the lawsuit against OpenAI is to demand that they change their name from OpenAI to Closed AI. OpenAI's original vision was that everything would be open-source, but now everything is closed. Other large AI labs, like Anthropic, are also completely closed. In fact, they have even stopped publishing research papers, treating everything as proprietary.

The DeepSeek team, for their own reasons, has actually fulfilled the promise of true open-source. They released the code for their LLM (called V3) and their reasoning engine (called R1), and published detailed technical papers explaining how they built it, essentially providing a roadmap for anyone else who wants to do similar work.

So it has been made public. There is a false narrative out there that if you use DeepSeek, you are giving all your data to the Chinese. This is true if you use the service on the DeepSeek website. But you can download the code and run it yourself. For example, Perplexity is an American company where you can use DeepSeek R1, fully hosted in the U.S. Microsoft and Amazon now have cloud versions of DeepSeek, and you can run it on their cloud platforms, both of which are American companies using American data centers.

This is very important. You can now download this system and actually run it on hardware worth $6,000 at home or in the office. Its capabilities are comparable to the cutting-edge systems of companies like OpenAI and Anthropic.

These companies have invested a lot of money to build their systems. Now, you can buy it for $6,000 and have complete control. If you run it yourself, you have full control. You can see transparently what it is doing, you can modify it, and you can perform various operations on it.

It also has a very impressive feature called distillation. You can compress large models that require $6,000 hardware to create smaller versions of the models. People online have already created smaller versions of the models and optimized them so that you can run them on a MacBook or iPhone. While these versions are not as intelligent as the full version, they are still quite smart. You can create customized, distilled versions tailored to specific domains that perform excellently in those areas.

This is a huge advancement in making large model reasoning and R1 model reasoning more accessible in programming and science. Just six months ago, these were still very arcane, extremely expensive, and proprietary. Now, it has become free and permanently available to everyone.

Every major tech company, internet company, and every startup—this week we have dozens or even hundreds of startups—are either rebuilding based on DeepSeek, integrating it into their products, or studying the technology they use to improve existing AI systems.

Mark Zuckerberg from the Meta team recently mentioned that the Meta team is dissecting DeepSeek and legally borrowing these ideas because it is open-source, ensuring that the next version of Llama will be at least comparable to DeepSeek in reasoning capabilities, if not better. This is indeed pushing the world forward.

The two main takeaways we can learn from this are: AI will be ubiquitous. There are many AI risk control people, security personnel, regulators, officials, governments, the EU, the British, and so on… all of these people want to limit and control AI, and this basically guarantees that none of that will happen, which I think is good. It aligns very well with the free tradition of the internet. Then, this achieves a 30-fold reduction in reasoning costs.

Perhaps it is worth noting that this indicates reasoning will work. Reasoning will work in any area of human activity as long as you can generate answers that can be checked for correctness by technical experts afterward.

We will have AI capable of human and superhuman level reasoning that will play a role in truly important fields: coding, mathematics, physics, chemistry, biology, economics, finance, law, and medicine.

This essentially guarantees that within five years, everyone on Earth will have access to a superhuman-level AI lawyer and AI doctor, always on standby, just a standard feature on their phones. This will make the world a better, healthier, and more wonderful place.

Patrick: But this is also the most unstable, as models can become outdated in just two months. There is a lot of innovation happening at every technological level. But just from this point in time, entering this new paradigm, if you were writing a column about the winners and losers among all stakeholders, whether new application developers, existing software developers, infrastructure providers like Nvidia, open-source versus closed-source model companies, who do you think are the winners and losers after the R1 release?

Marc: If you take a "snapshot" today, then from a zero-sum game perspective, the winners at a point in time are all the users, all consumers, every individual, and every business using AI.

There are some startups, like those providing AI legal services, that last week had costs of using AI that were 30 times what they are now.

For example, for a company building an AI lawyer, if the cost of its key inputs has dropped by 30 times, it’s like the cost of gasoline dropping by 30 times while driving. Suddenly, you can drive 30 times further on the same dollar, or you can use the extra spending power to buy more. All these companies will either greatly expand their ability to use AI in all these areas or they will be able to provide services in a cheaper or free manner. So for users and the world, this is a fantastic outcome based on a fixed-size pie.

The losers are those companies with proprietary models, like OpenAI, Anthropic, and so on. You will notice that OpenAI and Anthropic have issued quite strong but seemingly provoked messages in the past week, explaining why this is not the end for them. There is an old saying in business and politics, when you are explaining, you are losing.

Then another loser is Nvidia. There has been a lot of commentary on this, but Nvidia makes the standard AI chips that people use. There are some other options, but Nvidia is what most people use. Their chips have profit margins of up to 90%, and the company's stock price reflects that. (Nvidia) is one of the most valuable companies in the world. One of the things the DeepSeek team did in their paper was to figure out how to use cheaper chips, actually still using Nvidia's chips but using them more efficiently.

Part of the 30-fold cost reduction is that you need fewer chips. By the way, China is building its own chip supply chain, and some companies are starting to use China-derived chips, which is certainly a more fundamental threat to Nvidia. So this is a snapshot at a point in time. But the question is, your question implies another way to look at it, which is over time, what you want to see is the elasticity effect. Satya Nadella used the phrase "Jevons Paradox" to describe this.

Imagine gasoline. If the price of gasoline drops significantly, then suddenly people will drive more cars. This often occurs in traffic planning. So you would have a city like Austin that is congested, and someone might have the bright idea to build a new highway next to the existing one. And within just two years, the new highway will also be congested, perhaps even making it harder to get from one place to another. The reason is that the reduction in the price of key inputs can induce demand.

If AI suddenly becomes 30 times cheaper, people might use it 30 times more, or by the way, they might use it 100 times or even 1,000 times more. This economic term is called elasticity.

So a price drop equals explosive growth in demand. I think there is a very reasonable scenario here that on the other side, with the explosive growth in usage, DeepSeek will do very well. By the way, OpenAI, Anthropic will also do well, Nvidia will do well, and Chinese chip manufacturers will do well.

Then you will see a tidal effect, and the entire industry will experience explosive growth. We are really just beginning to figure out how to use these technologies. Reasoning has only started to work in the past four months. OpenAI only released their o1 reasoning model a few months ago. It’s like taking a spark from the mountain and handing it to all of humanity. And most of humanity has not yet used fire, but they will.

Then, frankly, this is also an old notion of creativity, which is to say, well, if you are OpenAI or a similar company, what you did last week is no longer good enough. But that said, that’s just the way the world works. You have to get better. These things are all competitions. You have to evolve. So this is also a very powerful catalyst that pushes many existing companies to truly raise their game and become more aggressive.

……

Patrick: …… If a Chinese company uses models developed in the U.S., which have been heavily invested in, and then leads to this rich technology for the world, it’s a difficult thing to understand. I would love to hear your response from these two perspectives.

Marc: Yes, so there are some real issues here. There is a certain irony in this argument, and you do hear this argument. Of course, the irony is that OpenAI did not invent the Transformer. The core algorithm of large language models is called the Transformer.

It was not invented at OpenAI; it was invented at Google. Google invented it and published the relevant papers, and by the way, they did not commercialize it. They continued to research it but did not commercialize it because, for "safety" reasons, they thought it might be unsafe. So they left it on the shelf for five years, and then the OpenAI team figured it out, picked it up, and continued to advance it.

Anthropic is a spinoff of OpenAI. Anthropic also did not invent the Transformer. So whether it is these two companies or every other American lab researching large language models, every other open-source project is built on things they did not create or develop themselves.

By the way, Google invented the Transformer in 2017, but the Transformer itself is based on the concept of neural networks. The idea of neural networks can be traced back to 1943. So, 82 years ago was actually when the original neural network paper was published, and the Transformer was built on 70 years of research and development, most of which was funded by the federal government and European governments in research universities.

Therefore, this is a very long lineage of knowledge and development, and most of the ideas that went into all these systems were not developed by the companies currently building these systems. No company is sitting here, including our own, with any special moral claim that we are building from scratch and should have complete control. That is simply not the case.

So, I would say that arguments like this stem from current frustrations. By the way, these arguments are also meaningless because China has already done this; it has come out, and things have happened. Now there is a debate about copyright. If you talk to experts in this field, many have been trying to understand why DeepSeek is so outstanding. One theory, which is an unproven theory but one that experts believe, is that Chinese companies may have used data that American companies did not use for training.

It is particularly surprising that DeepSeek is very good at creative writing. DeepSeek may currently be the best AI in the world for English creative writing. This is a bit strange because the official language of China is Chinese. While there are some very excellent Chinese English novelists, generally speaking, you might think that the best creative writing should come from the West. And DeepSeek may currently be the best, which is shocking.

So one theory is that DeepSeek may have trained on, for example, some websites called Libgen, which are basically huge internet repositories full of pirated books. I personally would not use Libgen, but I have a friend who uses it frequently. It is like a superset of the Kindle store. It has every digital book available in PDF format that you can download for free. It is like the Pirate Bay for movies.

American labs might not feel they can simply download all the books from Libgen and train on them, but perhaps Chinese labs feel they can. So there may be this differential advantage. That said, there is also an unresolved copyright dispute here. People need to be careful with this issue because there is an unresolved copyright dispute, and some publishing companies basically want to prevent generative AI companies like OpenAI, Anthropic, and DeepSeek from using their content.

There is an argument that these materials are copyrighted and cannot be used freely. There is another argument that basically says training AI on books does not mean you are copying the books; you are reading the books. AI reading books is legal.

You and I are allowed to read books, by the way. We can borrow books from the library. We can pick up a friend's book. These actions are all legal. We are allowed to read books, we are allowed to learn from books, and then we can go on with our daily lives, discussing the ideas we learned from the books. Another argument is that training AI is more like humans reading books rather than stealing.

Then there is the practical reality that if… their AI can be trained on all the books, and if American companies are ultimately legally prohibited from training on books, then America might lose the race in AI.

From a practical standpoint, this could be a fatal blow, like they won and we lost. There may be some entanglement in the entire argument. DeepSeek has not disclosed the data they used for training. So when you download DeepSeek, you do not get the training data; you get what is called the weights. Therefore, you get a neural network trained on the training materials. But it is difficult, if not impossible, to look into that and deduce the training data.

By the way, Anthropic and OpenAI also have not disclosed the data they used for training. Then there is intense speculation in the field about what is and is not in OpenAI's training data. They consider it a trade secret. They will not make this information public. Therefore, Chinese DeepSeek may be different from these companies, or it may not be. They may have different training methods than Chinese companies. We do not know.

We do not know what OpenAI and Anthropic's algorithms are because they are not open-source. We do not know how much better or worse they are compared to the publicly available DeepSeek algorithm.

Discussing Closed Source vs. Open Source

Patrick: Do you think those in the competition with closed-source models, like OpenAI and Anthropic, will ultimately become more like Apple versus Google's Android?

Marc: I support maximizing competition.** By the way, this aligns with my identity as a venture capitalist. So if you are a company founder, if I run an AI company, I need to have a very specific strategy that has pros and cons and requires trade-offs.

As a venture capitalist, I do not need to do that. I can make multiple contradictory bets. This is what Peter Thiel refers to as certainty optimism versus uncertainty optimism. Company founders and CEOs must be certainty optimists. They must have a plan and must make tough trade-offs to execute that plan. Venture capitalists are uncertainty optimists. We can fund a hundred companies with a hundred different plans and contradictory assumptions.

The nature of my work is that I do not need to make the kind of choices you just described. And that makes it easy for me to make a philosophical argument that I personally sincerely agree with, which is that I support maximizing competition. So, to go a layer deeper, this means I support free markets, maximizing competition, and maximizing freedom.

Essentially, if as many smart people as possible can come up with as many different methods as possible and compete with each other in a free market, let’s see what happens. Specifically regarding AI, this means I support large labs developing as quickly as possible.

I 100% support OpenAI and Anthropic doing whatever they want, launching any products they want, and developing as hard as they can. As long as they do not receive preferential policy treatment, subsidies, or support from the government, they should be able to do whatever they want as companies.

Of course, I also support startups. We are certainly actively funding AI startups of various sizes and types. So, I hope they can grow, and then I hope open source can grow, partly because I believe that if things appear in open source, even if it means that some companies with business models cannot operate, the benefits to the world and the entire industry are so great that we will find other ways to make money. AI will become more ubiquitous, cheaper, and easier to access. I think that will be a good outcome.

Then, another very critical reason for open source is that without it, everything becomes a black box. Without open source, everything becomes a black box owned and controlled by a few companies that may ultimately collude with the government, and we can discuss that. But you need open source to be able to see what is happening inside the box.

By the way, you also need open source for academic research, so you need open source for teaching. So the problem before open source was, going back two years, when there were no basic open-source LLMs, Meta released Llama, then France's Mistral, and now DeepSeek.

But before these open-source models appeared, the university system was experiencing a crisis in that university researchers at places like Stanford, MIT, and Berkeley did not have enough funding to purchase billion-dollar Nvidia chips to truly compete in the AI field.

So if you talked to computer science professors two years ago, they would be very concerned. The first concern was that my university does not have enough funding to compete in the AI field and remain relevant. Then another concern was that all universities combined also do not have enough funding to compete because no one can keep up with the fundraising capabilities of these large companies.

Open source has brought universities back into the competition. This means that if I am a professor at Stanford, MIT, Berkeley, or any state school, whether it is the University of Washington or elsewhere, I can now use the Llama code, Mistral code, or DeepSeek code for teaching. I can conduct research, and I can actually make breakthroughs. I can publish my research findings and let people truly understand what is happening.

Then, every new generation of kids coming to university, taking computer science courses, will be able to learn how to do this, and if this is a black box, they cannot do it. We need open source just as we need freedom of speech, academic freedom, and research freedom.

So my model is basically that you let big companies, small companies, and open source compete with each other. That is what has happened in the computer industry. It works very well. That is what has happened in the internet industry. It works very well. I believe this will happen in the AI field, and I think it will work very well.

Patrick: Is there a limit to wanting to maximize the speed of evolution and the level of competition? Maybe there is. If I say we know the best things are made in China, …, is there a situation where you say, yes, I want to maximize evolution and competition, but national interests somehow outweigh the desire for maximum speed of evolution and development?

Marc: This argument is a very real argument. It is frequently raised in the AI field. In fact, as we sit here today, there are two things. First, there are currently actual restrictions on Western companies and American companies selling cutting-edge AI chips to China. For example, Nvidia cannot legally sell its cutting-edge AI chips to China today. We live in a world that has already made this decision and implemented this policy.

Then the Biden administration had issued an executive order, which I believe has now been rescinded, but they had issued an executive order that would impose similar restrictions on software. This is a very active debate. With the occurrence of the DeepSeek event, Washington, D.C. is undergoing another round of such debates.

Then basically, when you get into policy debates, you encounter a classic situation where you have a rational version of the argument, which is theoretically what is in the national interest. Then you have a political version of the argument, which is, well, what will the political process actually do to the rational argument? Let me put it this way: we all have a lot of experience watching rational arguments collide with the political process, and usually, it is not the rational argument that wins. The outcome that comes out of the political machine is often not what you initially thought you would get.

Then there is a third factor that we always need to discuss, which is particularly the corrupting influence of large companies. If you are a large company and you see the changes happening with Chinese companies (becoming more competitive) and the threat of open source, of course, you would try to leverage the U.S. government to protect yourself. Maybe that aligns with the national interest, maybe it doesn’t. But you will certainly push for it, regardless of whether it aligns with the national interest. This is what complicates the debate.

You cannot sell cutting-edge AI chips to China. This certainly hinders them in some ways. There are some things they will not be able to do. Maybe that is a good thing because you have decided that this is in the national interest. But let’s look at the three other interesting consequences that arise from this.

So one consequence is that it provides a huge incentive for Chinese companies to figure out how to achieve things on cheaper chips. This is a significant part of DeepSeek's breakthrough, as they figured out how to use legal, compliant, cheaper chips to do what American companies could only do with larger chips. This is also one reason why it is so cheap. One reason is that you can run it on hardware worth $6,000 because they have invested a lot of time and effort into optimizing the code to run efficiently on cheaper chips that are not sanctioned. You have forced an evolutionary response.

So that is the first reaction, and perhaps it has backfired to some extent. The second consequence is that you incentivize both the state-owned and private sectors in China to develop a parallel chip industry. So if they know they cannot get American chips, then they will go develop their own. They are doing that now. They have a national plan to build their own chip industry so that they do not rely on American chips anymore.

So from a counterfactual perspective, maybe they would have purchased American chips. Now they will figure out how to manufacture them themselves. Maybe in five years, they will be able to do that. But once they reach a point where they can manufacture on their own, then we will have a direct competitor in the global market that we would not have if we were just selling them chips. And by the way, by that time, we will have no control over their chips. They can have complete control. They can sell them at below-cost prices; they can do whatever they want.

How AI Reasoning Capabilities Change VC and Investment Industry

Patrick: How do you think all of this will affect capital allocation? I am most interested in how, perhaps five years from now, your company, Andreessen Horowitz (A16Z), will be affected. If I think of investment firms as a combination of those who can raise capital, do great analytical work, and have the ability to judge people, especially in early-stage investments, how do you think this function will change with the emergence of “o7” (AI reasoning capabilities)?

Marc: I hope the analytical part can undergo a huge transformation. We assume that the best investment firms in the world will be very good at leveraging this technology to conduct the analytical work they do.

That said, there is a saying that “the shoemaker's son has no shoes,” perhaps the venture capital firms that are most aggressively investing in AI may be among those that are not aggressive enough in practical applications. But we are undertaking several efforts internally at our company, and I am very excited about that. But companies like ours need to keep up with the times, so we really have to do this.

Is there some work already underway within the industry? Perhaps not yet. Perhaps not enough. That said, for late-stage investments or public market investments, many people we talk to have a very analytical perspective. There are even great investors, I think of Warren Buffett. I don’t know if this is true, but I have always heard that Warren never meets with CEOs.

Patrick: He wants “ham sandwich companies.”

Marc: Yes, yes, he wants companies that are as simple as a ham sandwich. And I think he is a bit concerned that he might be swayed by a good story. You know, many CEOs are very charismatic people. They are always described as “having great hair, white teeth, shiny shoes, and sharp suits.” They are very good at sales. You know, one of the things CEOs are good at is selling, especially selling their own stock.

So if you are Buffett, you are sitting in Omaha, and what you do is read annual reports. Companies list everything in their annual reports, and they are bound by federal law to ensure that the content is truthful. So that is your way of analyzing. Now, do reasoning models like o1, o3, o7, or R4 do better than most investors manually analyzing annual reports? Perhaps they do.

As you know, investing is an arms race, just like everything else. So if it works for one person, it will work for everyone. It will become an arbitrage opportunity for a while, and then it will close and become standard. Therefore, I expect the investment management industry to adopt this technology in this way. It will become a standard operating procedure.

I think the situation is a bit different for early-stage venture capital. What I am about to say may just be my personal wishful thinking. I might be the last Japanese soldier on a remote island in 1948 saying what I am about to say. I am going to take a risk. But I want to say that, look, in the early stages, a lot of what we do in the first five years is actually deeply assessing individuals and then working very closely with those people.

This is also why venture capital is hard to scale, especially geographically. Geographical scaling experiments often do not work. The reason is that you ultimately need to spend a lot of time face-to-face with these people, not just during the assessment process but also during the building process. Because in the first five years, these companies are usually not yet in autopilot mode.

You actually need to work closely with them to ensure they have everything they need to succeed. There is a very deep interpersonal relationship, dialogue, interaction, and guidance, by the way, we learn from them, and they learn from us. It is a two-way exchange.

We do not have all the answers, but we have a perspective because we see a broader panorama, while they are more focused on specific details. Therefore, there is a lot of two-way interaction. Tyler Cowen talked about this, and I think he referred to it as “project selection.”

Of course, “talent scouting” is another version, which is basically, if you look back at any new field in human history, you can almost always find this phenomenon, where there are some uniquely individual people trying to do something new, and then there is a professional support layer that funds and supports them. In the music industry, David Geffen discovered all the early folk artists and turned them into rock stars. Or in the film industry, it was David O. Selznick who discovered early movie actors and turned them into movie stars. Or in a café or tavern in Maine 500 years ago, people were discussing which whaling captain could go catch whales.

You know, this is Queen Isabella listening to Columbus's proposal in the palace and saying, “Sounds reasonable. Why not?” This alchemy that has developed over time between those doing new things and the professional support layer that supports and funds them has existed for hundreds, if not thousands, of years.

You might have seen tribal leaders thousands of years ago sitting around a fire, with a young warrior coming up and saying, “I want to lead a hunting party to that area over there to see if there are better game.” And the leader sitting by the fire trying to decide whether to agree. So this is a very human interaction. My guess is that this interaction will continue. Of course, that said, if I encounter an algorithm that is better at doing this than I am, I would retire immediately. We shall see.

Patrick: You are building one of the largest companies in this field. How do you adjust the company’s development strategy to respond to this new technology? Have you made adjustments in terms of practical operations or strategic direction? How do you adjust the company’s development direction to respond to this new technology?

Marc: An important part of running a venture capital firm, in our view, is having a set of values and behaviors that you must possess, which we call timeless. For example, respect for entrepreneurs. You need to show great respect for entrepreneurs and the journeys they go through. You need to deeply understand what they are doing. You cannot just skim the surface.

You need to build deep relationships. You need to work with these people over the long term, and by the way, these companies take a long time to build. We do not believe in overnight success. Most great companies are built over spans of 10, 20, or 30 years. Nvidia is a great example. Nvidia is about to celebrate its 40th anniversary, and I think one of Nvidia's original venture capitalists is still on the board today. This is a great example of long-term building.

So, there is a core set of beliefs, perspectives, and behaviors that we will not change, which relate to what we just mentioned. Another is the face-to-face communication aspect. You know, these things cannot be done remotely, that is one thing. But on the other hand, you need to keep up with the times because technology is changing so rapidly, business models are changing so quickly, and competitive dynamics are changing so fast.

If anything is different, the environment has become more complex because now you have many countries, and now there are all these political issues, which also complicates things. We never really worried about the political system putting pressure on our investments until about eight years ago. Then about five years ago, that pressure really intensified. But in the first ten years of our company and the first 60 years of venture capital, this was never a big deal, but now it is.

Therefore, we need to adapt. We need to engage in politics, which we did not do before. Now we need to adapt, and we need to figure out that maybe AI companies will be fundamentally different. Maybe their organizational structures will be completely different. Or as you said, maybe the way software companies operate will be entirely different.

We often ask ourselves a question, for example, what does the organizational structure of a company that fully leverages AI look like? Is it similar to existing organizational structures, or will it actually be very different? There is no single answer to this, but we are seriously thinking about it.

So, one subtle balancing act we do every day is trying to figure out what is timeless and what needs to keep up with the times. Conceptually, this is an important part of my thinking about the company, that we need to navigate between these two and ensure we can distinguish between them.

Patrick: Your company is now quite large, and in some ways, it resembles companies like KKR or Blackstone. You and Ben Horowitz, as founders, are both experienced founders. When you started this company, similar to Blackstone, Stephen Schwarzman had never really made investments before founding Blackstone. Look at its development now.

It seems that this founder-led approach to building asset management investment firms ultimately evolves into truly large and ubiquitous platforms. You have vertical businesses that cover most of the exciting cutting-edge technology areas. Do you think there is some truth to this perspective? Will the best capital allocation platforms be founded more by founders rather than investors?

Marc: Yes, so there are a few points. First, I think this observation has some validity. Within the industry, people often talk about how many investment operations are typically referred to as partnerships. Many venture capital firms operate this way. Historically, it has been a small team of people sitting in a room, exchanging ideas, and then making investments. By the way, they do not have a balance sheet. It is a private partnership. They are compensated at the end of each year. This is the traditional venture capital model.

A traditional venture capital model has six general partners (GPs) sitting around a table operating this way. They have their assistants and a few associates. But the point is, it is entirely people-based. By the way, you will actually find that in most cases, people do not really like each other.

"Mad Men" illustrates this well. Remember in "Mad Men," in the third or fourth season, members left to start their own companies, and they did not actually like each other. They knew they needed to come together to start a company. This is how many companies operate. So, it is a private partnership, and that is what it represents.

But then you see that these companies struggle to sustain themselves. They lack brand value. They do not have potential enterprise value. They are not a business. The companies you see in this model are those where the original partners, when they are ready to retire or do something else, pass it on to the next generation. Most of the time, the next generation cannot continue to sustain it. Even if they can sustain it, there is no potential asset value. The next generation will have to pass it on to the third generation. It may fail in the third generation, and then it will eventually appear on Wikipedia. It will say something like, “Yes, this company once existed, and then it disappeared, replaced by others, like ships passing in the night.”

So this is the traditional way of operating. By the way, if you have received traditional investment training, you have been trained in the investment part, but you have never been trained on how to build a business. So, it is not your natural strength; you do not have that skill or experience, so you do not do it. Many investors have operated as investors this way for a long time and made a lot of money. So, it can work well.

Another way is to build a company, establish a business, and create something with lasting brand value. You mentioned companies like Blackstone and KKR, these huge publicly traded companies. Apollo is similar; these massive companies—you may know that the original banks were actually all private partnerships. Goldman Sachs and JPMorgan a hundred years ago were more like today’s small venture capital firms than they are now. But then, their leaders transformed them into these massive enterprises over time. They are also large publicly traded companies.

So, this is another way, to establish a franchise. Now, to do this, you need a theory of why a franchise should exist. You need a conceptual theory of why doing this makes sense. Then, yes, you need business skills. By that time, you are running a business, just like running any other business, meaning, okay, I have a company. It has an operating model, it has an operational rhythm, it has management capabilities, it has employees, it has multiple layers, and it has internal specialization and professionalization.

Then you start thinking about scaling, and over time, you begin to consider potential asset value, meaning the value of this thing is not just in the people currently there. It is not like us, eagerly wanting to distribute profits or anything else. But one big thing we are trying to do is build something with this kind of permanence.

By the way, we are not in a hurry to go public or anything else, but one big thing we are trying to do is build something with this kind of permanence.

Patrick: What new differences do you hope the company will have in the next ten years that do not currently exist? Are there some uncompromising ways you hope the company will never evolve like traditional large asset management companies?

Marc: We are rapidly evolving in terms of the investment objects, what the company does, the models, and the backgrounds of the founders. For example, there has been a consensus in the venture capital world for 60 years that you would never support researchers starting companies to conduct research. They would just do research, burn through funds, and in the end, you would get nothing.

However, many of today’s top AI companies are indeed founded by researchers. This is an example of how some so-called “timeless” values need to be adjusted according to the times. We need to maintain a high degree of flexibility regarding these changes. Therefore, as these changes occur, the help and support needed for the company’s success will also change accordingly.

One of the most significant changes in our company, which I mentioned before, is that we now have a large and increasingly complex political operations department. Four years ago, we had a blank slate in the political realm. Now, it has become an important part of our business, something we had never anticipated before.

I am confident that in ten years, we will not only invest in areas that are currently unimaginable but also have operational models that are currently unimaginable. Therefore, we are completely open to changes in these aspects. However, there are some core values that I hope will remain unchanged over the next ten years because these values have been thoughtfully considered and are the foundation of our company.

But I have always emphasized to our team members and limited partners that we are not scaling for the sake of scaling. Many investment firms, once they reach a certain size, prioritize expanding their asset management scale from billions to hundreds of billions or even trillions of dollars. This practice is often criticized for focusing more on collecting management fees rather than achieving excellent performance in investments. That is not our goal.

The only reason we scale is to support the companies we want to help founders build. When we scale, it is because we believe it helps us achieve that goal.

However, I must emphasize that the core of our company has always been early-stage venture capital. No matter how large we become, even if we establish growth funds that can write larger checks—some AI companies indeed require significant funding. We did not set up growth funds from the beginning; they were gradually established as market demand and company development evolved.

But the core business has always been early-stage venture capital. This may be confusing because, from the outside, we manage a large amount of capital. Why would I, as a founder of an early-stage startup, believe that you would be willing to spend time on me? Because you, Andreessen Horowitz, have invested hundreds of millions in late-stage investments, while you only invested $5 million in my Series A. Will you still spend time focusing on me?

The reason is that the core business of our company has always been early-stage venture capital. From a financial perspective, the return opportunities in early-stage investments are comparable to those in later-stage companies, which is characteristic of startups. But more importantly, all of our knowledge, network of relationships, and what makes our company unique comes from our deep insights and connections in the early stages.

So, I always tell people that if circumstances force us, and the world is in trouble, and we must make sacrifices, then the early-stage venture capital business will never be sacrificed. This will always be the core of the company. That is also why I spend a lot of time working with early founders. On one hand, it is very interesting; on the other hand, it is also where I learn the most.

The Shift in Global Power Structures: Elites vs. Anti-Elites

Patrick: If we consider the changes in global power structures, …, which power centers are you most concerned about that are either gaining or losing power?

Marc: "The Machiavellians." I am sure you have probably had a dozen people recommend this book on your show. It is one of the greatest books of the 20th century. It articulates theories about political power, social, and cultural power. There is a key point in this book that I see everywhere right now, which is the concept of elites and anti-elites.

The idea is this: basically, democracy itself is a myth. You will never have a fully democratic society. By the way, the United States is certainly not a democratic country; it is a republic. But even those “democratic” systems that operate well tend to have a republican nature, with a lowercase “r.” They tend to have a parliament, or a house of representatives and a senate, or some kind of representative body. They tend to have a representative institution.

The reason is a phenomenon described in this book called the “oligarchic iron law,” which is basically this: the problem with direct democracy is that the masses cannot organize. You cannot really get 350 million people to organize to do anything. There are too many people.

So, basically, in every political system throughout human history, you have a small, organized elite governing a large, disorganized mass of the populace. From the earliest hunter-gatherer tribes to every other political system in modern times, whether the Greeks or Romans, or every empire and nation in history.

So, a small, organized elite governs a large, disorganized mass of the populace. This relationship is fraught with danger because the disorganized masses may comply with the elite for a time, but not necessarily forever. If the elite becomes oppressive to the masses, the number of the masses far exceeds that of the elite. At some point, they may show up with torches and pitchforks. So, there is tension in this relationship. Many revolutions occur because the masses decide that the elite no longer represents them.

Our society is no exception. We have a large, disorganized mass of the populace. We have a very small, organized elite. The United States… has established a system where we have two elite classes. We have the elite class of the Democratic Party and the elite class of the Republican Party. By the way, there is a significant overlap between these two elite classes, and some people actually refer to it as a “single party.” Perhaps there are more commonalities between these elite classes than between them and the masses.

For a long time, we have had a Republican elite whose policies are ultimately represented by the Bush family. We have a Democratic elite whose policies are ultimately represented by Obama. Over the past decade, there has essentially been a rebellion within the elite classes on both sides in the United States. This is actually a key point in "The Machiavellians," which is that change often does not come from the masses directly confronting the elite. What happens is the emergence of a new anti-elite class.

You will have a new anti-elite class emerging, trying to replace the current elite class. My interpretation of current affairs is that, generally speaking, the elite class running the world today is found to be doing poorly. We can discuss the reasons later. But generally, if you look at the approval ratings of (Western) political leaders and institutions, all of these are declining. What is happening everywhere in the world is that if you are an incumbent institution, if you are an incumbent newspaper, if you are an incumbent television network, if you are an incumbent university, if you are an incumbent government, generally speaking, your public support ratings are a disaster. This is basically what people are saying: the ruling elite class is failing us.

Then these anti-elite classes emerge, saying, “Oh, I know I have a better way to represent the masses, I have a better way to take over.” My new anti-elite movement should replace the current elite movement, such as in the case of the Democratic Party. In 2016, this was Bernie Sanders, and now it is Alexandria Ocasio-Cortez (AOC) and the whole progressive wave. On the Republican side, this is clearly Trump and his “Make America Great Again” (MAGA) movement and everything it represents.

By the way, this dynamic is also happening in the UK. The Conservative Party has collapsed, and now you have this Reform Party with Nigel Farage, which is very threatening. You have Jeremy Corbyn, who is also an anti-elite figure from the left.

The same is true in Germany. In fact, just this week, something very dramatic happened in Germany, where the so-called “far-right” party AfD is rising rapidly. There is a leader named Alice Weidel, and for the first time in German political history, the Christian Democratic Union (CDU) has actually cooperated with the AfD on something after 50 years or more. Suddenly, the AfD has become a viable competitor. They are an anti-elite class trying to take over the German political system from the right.

So, basically, no matter where you go in the world, there is an anti-elite class emerging, saying, “Oh, I can do better.” This is a struggle between elite classes. The masses are aware of this, and they are watching democratic societies, and they will ultimately make a decision because they will decide who they want to vote for.

That is why Republican voters decided to vote for Trump instead of Jeb Bush. This is the case of the anti-elite class defeating the elite class. This is also related to the criticism of Trump, which is very interesting, that Trump is criticized by the existing elite class saying, “Oh, he is not one of the people. He is a super-rich billionaire who lives in a golden penthouse, and he has people driving him everywhere. If you are a rural farmer in Kentucky or Wisconsin, you shouldn’t think of him as one of you.”

The point has never been that Trump is one of the people. The point is that Trump is an anti-elite figure who can better represent the people. This is the foundation of his entire movement. By the way, the media space is the same. Everything you described is exactly what is happening in the media space. Elite media has dominated for 50 years; it is television news, cable news, newspapers, and these well-known magazines. Now you have the anti-elite class. The anti-elite class is you, Patrick, and (famous podcast host) Joe Rogan. And there are more people.

By the way, if you look at the numbers, it is very clear that the masses, the audience, and the readers are leaving the old media and turning to the new media. The existing elite class is very angry about this. They angrily write all these negative articles about you guys, saying you are a bunch of white supremacists, and the whole thing is terrible. It’s like, this is the way the world is. So we are in the midst of all this. I don’t know if “transition” is the right term. It is more like an intense battle between the old elite class and the new elite class.

Patrick: What are the initial seeds that led to the decline of the previous generation of elites, resulting in those 11% approval ratings? What do you think this is primarily attributed to?

Marc: There are two theories. One theory is that these approval ratings are wrong, and the other theory is that these approval ratings are correct. By “wrong,” I mean that these approval ratings are measured correctly, but people are giving the wrong answers.

If you are the head of CNN or Harvard University, or you are responsible for any similar institution, and your approval rating is only 11%… By the way, Gallup has been conducting a remarkable survey for 50 years called “Institutional Trust.” You can Google “2024 Gallup Institutional Trust Survey,” and you will see some spectacular charts showing that institutional trust peaked in the late 1960s and early 1970s and has been declining ever since.

By the way, this phenomenon predates the advent of the internet. Interestingly, it has been blamed on the internet, but it predates the internet. So, this is a phenomenon that has been developing since the 1970s and has been accelerating. By the way, since 2020, the rate of decline in these approval ratings has been even faster.

They slide down like this, and then after 2020, they drop straight down. Television network news, I don’t know what the specific numbers are. It is in the single digits; people completely no longer trust it. They no longer believe what is said on television news. By the way, audience ratings are also declining in the same way.

So, one theory is that if you are the head of NBC News or CNN or Harvard University, your theory might be, “Oh, people are wrong. People are misled; they are deceived; they are fooled by populists and demagogues; they are deceived by false information.” This is why the concept of “fake news” has become so popular. … People are deceived by malicious actors, populists, and demagogues; it is just a matter of time until we explain to people that they have been deceived. They will believe us again.

So, that is one theory. The other theory is that the elite class has become corrupt. They have become corrupt, dysfunctional, and corrupt; they are no longer providing service. Under this theory, these numbers, the decline in approval ratings, are correct because every time you see Congress, they are shamelessly spending your money on all sorts of crazy things. If you look at CNN or NBC News, they are always lying to you about a thousand different things. If you go to Harvard, they will teach you racial communism, that America is evil, and so on, all these crazy things.

Under this theory, people are correct; people have seen through these elite classes. These elite classes have basically been in power too long, they have too much power, they have not been sufficiently scrutinized, they have not faced enough competitive pressure, they have become corrupt in place, and they are no longer providing service. The reality may be that both situations exist. It is easy for the next demagogue to emerge and just start throwing stones at the powers that be, saying anything.

If you are someone today who has no political power but wants it, the easiest thing to do is to show up and start shouting that the current elite class is corrupt. Maybe that is a bit true; demagoguery works a bit, or whatever, but… I think the main reason is that the elite class has become corrupt.

My version is very straightforward; Burnham talks about this in his book. He talks about the “cycle of elites.” He says that for an elite class to remain healthy, real, productive, and not corrupt, it needs to constantly inject new talent. It does this through the process of the elite cycle.

So, what it will do is identify promising young talent and invite them to join the elite class. It does this for two reasons. One is for self-renewal. The other is that those people are most likely to become the anti-elite class. So, this is also to prevent future competition. So, my experience starting from when I was 22 is, “Oh, hey, Marc, we really want you to come to Davos. We really want you to come to Aspen. We really want you to come to New York for this big conference. We really want you to come to the New York Times dinner party. We want you to hang out with the journalists for 25 years.” That is what I did; it was like, “Oh, that sounds great. These are the best people in the world. They are in charge of everything. They have the best degrees; they graduated from the best schools. They hold all the power positions. They like me. They think I’m great.”

They keep praising me for coming from the cornfields of Wisconsin. I made it; I entered the elite class.

All I had to do was never argue with anything. All I had to do was agree with anything that was said in the New York Times, agree with anything said at Davos, vote for the candidates you should vote for, donate to the candidates you should donate to, and never, ever, ever stray off course. Then you become part of the elite class.

I have many peers who have done this. Some are now the largest Democratic donors in the world; they have completely integrated into the elite class, they are there, they are having a great time, and they think it is all wonderful, it is great. Some people feel good about it; maybe that is the right thing to do.

Then some people, at some point, look around. It’s like the story of J.D. Vance. He grew up in rural Kentucky or the Appalachian region of Ohio. He eventually went to Yale University. He eventually got invited into all these inner circles.

Then he eventually looked around, and he just said: “Wow, these people are not at all what I imagined. These people are selfish, corrupt, they are lying about everything, they are engaging in speech suppression, they are very authoritarian, and they are plundering the public treasury. Oh my God, I have been deceived my whole life. These people do not deserve the respect they have; maybe there should be a new elite class in power.” So, this is a lot of the debate that is unfolding right now. Yes, I am a case study.

Optimism vs. Pessimism: Will the World Be Better?

Patrick: If we put on a pair of optimistic glasses, you emphasize early-stage venture capital. You will encounter all these young, smart people who are about to build the future. Let’s put on a pair of optimistic glasses and assume that AI has the most positive impact in all areas where we can verify results. Reasoning becomes so powerful.

So, what related bottlenecks might hinder the expected technological revolution? It could be clinical trials in medicine, or the progress of certain things is slower than AI, and AI is not the problem. We will be eager to make progress.

But the atomic world, the surveillance world, or the world of clinical trials, etc., may become limiting factors rather than intellect and knowledge. What bottlenecks are you most interested in?

Marc: The way I have always thought about technological change is that there used to be three lines on the chart, and now there are four lines. So, one line is the speed of technological change, which is one line, and everything is generally getting better and better. Then every once in a while, you see these discontinuous leaps, or something becomes dramatically better, just like what happened with AI last week.

Then you have another line above it, which is social change, basically when the world is ready to accept new things. Sometimes you see this phenomenon where new things actually exist before the world is ready for them, and for some reason, they are not adopted. Then five years or fifty years later, they suddenly take off and develop rapidly. So, there is a social level, and above that, there is a financial level, which is whether the capital markets are willing to fund it. Can it generate returns?

I think the art of being an entrepreneur or a tech investor is trying to bridge these three.

So, you are trying to support something where the technology is truly ready, society is ready to adopt it, and you can actually get funding for it or take it public .

So, you have to align these three curves.

A lot of what we do in our daily work is aligning these three curves. A fourth line has emerged in the past five years. Over the past four years, the overwhelming answer has been the government. This is very strange and unsettling to me when I first encountered it because I am not used to it. And I have never seen us as being political or partisan, or that we are really trying to seek favor in Washington. We also haven’t tried to get subsidies. But we also didn’t think we needed to do anything to avoid being trampled. Then this situation suddenly happened.

……

Patrick: What do you feel most about the way this elite class wants to destroy you? How does it manifest?

Marc: This coincides roughly with a national shift in sentiment, possibly between 2013 and 2017. I grew up in the 90s, politically, I was a default Democrat for Clinton and Gore. At that time, there was a “Deal” (The Deal), capital D, that yes, you are a Democrat, but Democrats are pro-business, they love technology, they love startups. Clinton and Gore loved Silicon Valley. They loved new technology. They were always excited about what we were doing. They were always willing to help us if other countries came after us or whatever. They were always trying to help and support us.

Yes, you could be a pro-business, pro-technology Democrat. That was great. You could make a lot of money. People would write a lot of great articles about you, and then you donate all the money, you become a philanthropist, and that’s wonderful.

You die, and your obituary will say he was a great entrepreneur and a great philanthropist, and everything is beautiful. Basically, starting in 2013, every aspect of this deal collapsed. It manifested in many ways, but first and foremost in media coverage. The mainstream media’s official institutions began to turn against us, and everything we did was evil. This was actually quite surprising. In 2012, social media was seen by the mainstream media as an absolute, pure good because it helped Obama get re-elected, ….

Everyone knew it would only elect the right political candidates, … Then by 2016, the narrative completely reversed. Social media, along with the internet and technology, was destroying democracy, and everything was being undermined. So, media coverage was like the canary in the coal mine.

Part of the reason is that the employee base was radicalized, by the way. A strange situation arose where these large investment managers emerged, demanding that you take radical political stances in your company, which was completely absurd at the time. Then ultimately, the government itself emerged, and the bureaucracy of the Trump administration began to do this, which was beyond his direct control.

But under the Biden administration, this turned into an organized movement, which I would describe as destruction, accompanied by endless lawsuits, investigations, Wells notices, de-banking, censorship, attacks, trying to completely destroy entire industries. Of course, this is ultimately why we reacted. My hope is that all of this is over. That is to say, the new government is taking a very different approach and is no longer doing all these things.

Then my hope is that the next Democratic government will realize that attacking technology and attacking startups is actually unnecessary. In fact, it may be counterproductive because if you push Elon Musk out of your camp, there are consequences. I talk to many Democrats; we support many Democrats in the company, many members of Congress and senators, and I will talk to them again next week.

Basically, what they tell me is, look, there is a civil war within the Democratic Party, one side is us, who think the party should return to the center, stop attacking capitalism, attacking business, and attacking technology, and just win elections again.

Then there are some who believe the party actually needs to become more radical, we need to distinguish ourselves more from the other side, we need to become more extreme in economic policy, technology policy, and social policy. They are fighting for that. My hope is that they will return to the center so that we don’t have to go through all this again. We can maintain positive relationships with both sides, but we will see what happens.

Patrick: Like many others, I am very interested in the nature and state of global supply chains. When you delve into the components of pharmaceuticals or many other things, you see how interdependent the world is, especially the U.S. dependence on the external world for general supply chains.

I am curious how you think and hope this state will evolve in the next decade or so, because clearly, there are reasons we have gone global. But now, there are indeed many vulnerable links in global supply chains. How do you view the evolution of this part of the economy and economic narrative? Going back to what you just mentioned, you hope the U.S. wins supply chain manufacturing; how will the U.S. win this competition, and all these exciting ideas you hear today?

Marc: Yes, this is really important, and it is very different from the past. … As you know, the complexity of supply chains. Take the iPhone as an example; it is a typical product. There is a document you can download online, which may be a bit outdated. But it lists the components that make up the iPhone and where these components come from. A document I read ten years ago, there may be an updated version now, but the document I read ten years ago showed that at least at that time, the parts of the iPhone came from 40 different countries.

So, when the iPhone is assembled in a Foxconn factory in China, there are actually 39 countries that have already sent parts over, which are assembled into subcomponents of subcomponents and then become components. The same goes for cars; robots will be like this, anything complex, anything computerized or mechanical will have this property. By the way, this is actually hard to get from trade numbers because I believe this is correct.

China actually gets all the credit for the entire export value of the iPhone in the export numbers, even though the economic value added that occurs in China is actually a single-digit percentage. Because most of the stuff in the iPhone comes from the other 39 countries. What you really want to do is an analysis called economic value added analysis. You basically want to say, well, of the $1000 that goes into the iPhone, where does the value of these things come from, in dollars? The answer is from all over the world.

This is the issue of simply offshoring or reversing globalization; we are not talking about bringing steel mills back from China to the U.S. We are talking about unraveling a supply chain involving 40 countries, where things are moving back and forth because everything is being built and assembled. By the way, this is also a problem of the modern economy, which conflicts with reality in many ways.

……

Then there are political and economic pressures; the U.S. political system has assumed for 30 years that you can offshore manufacturing from the U.S., and those Midwestern and Southern communities that see all the factories closing will just sit by and think of other ways. In many places in the U.S., they never thought of new ways. It turns out they can still vote.

Part of the reason is that in my country (the U.S.), many people have been radicalized because the government and corporations seem to think it is okay to hollow out the economy and send everything overseas.

So, part of what is happening in the U.S. political system is that they have decided they will no longer accept this practice, and they will vote for something different. At the time, someone made this point, but the argument for economic efficiency won and brought benefits. It paid off in some ways. But many people in America have been radicalized. I come from a place where many people have been radicalized because the government and corporations seem to think it is okay to hollow out the economy and send everything overseas.

So, even if you gain returns from economic efficiency, your political system may not be able to bear it. You may very well regret it. I think there are no simple answers here. Anyone, in my view, who says there are simple answers here is wrong. It is complex.

The possible scenario is that the world will remain highly interdependent, with significant pressures and back-and-forth fluctuations. This dynamic will continue alongside tariffs and trade negotiations. It will be an ongoing process, with twists and turns along the way, but fundamentally, the world will remain interconnected in many ways, and we will manage to cope.

The question is, if at some point a war or a more severe pandemic occurs, or something similar, this interdependence could be severely stressed to the point of breaking. I hope that does not happen. But to some extent, the more interconnected the world is, the more resilient it is because there are more ways to do things, people have more ways to adapt, and everything can change. Then in some ways, the more interconnected the world is, the more dangerous it is because if one part breaks, the whole system can break. So, there is a real tug-of-war here.

Talking about Yushu and the Chinese Robotics Industry: "This specific environment is called Shenzhen"

Patrick: There is another area lurking at the forefront of technology that I haven’t seen you talk about much, which is robotics. Everyone is very excited about its potential. It is easy to imagine a humanoid robot that can do all the things humans no longer need to do. To make this world a reality, a lot of technological breakthroughs are needed. What do you think will happen in the field of robotics? What is overestimated? What is underestimated? How do you view it?

Marc: I would list four things. So, I would say smartphones, drones, cars, and robots. Basically, this is the ladder that China is climbing . By the way, this is not just about products, but the ladder of the entire supply chain. So, China has become the place where all the phones are assembled and manufactured. So, as you know, they have built a complete ecosystem in China, with thousands of specialized companies basically manufacturing various electronics and hardware, mechanical and computer-related things.

This specific environment is called Shenzhen, which is a cluster of thousands of companies that basically manufacture various electronics and hardware, mechanical and computer-related things. So, their phones’ (supply chain), and then they leverage this supply chain to win the drone market in China. Consumer drones, like DJI drones. Basically, China has won the global drone market, with a market share of over 99%.

……

In many ways, drones are like flying smartphones. They have many of the same devices, and then they have some new things, but they want to enter this field, at least until recently. Now they are entering the automotive field. **The reason is that a modern self-driving electric car is more like a laptop on wheels **** running ** or more like a smartphone on wheels, rather than a traditional internal combustion engine car.

Tesla in the U.S. is an example where Tesla is essentially a computer wrapped in many battery packs within a frame, with some tires on the outside. A good change illustration is if you go to the service area of a traditional car dealership compared to the service area of a Tesla dealership. The service area of the traditional automotive industry is filled with oil and dirt, everyone is in work uniforms, and they spend all day wiping their hands with a dirty rag.

You go to the service area of a Tesla dealership, and it’s like a surgical room. Everything is clean because it’s an electric vehicle, with no internal combustion engine. All that oil and dirt stuff is gone; it’s just a computer. The Chinese are basically doing in the automotive field what they once did in the drone and smartphone fields, which is they have built a complete ecosystem leveraging these other supply chains. They have established a complete ecosystem with all the components needed to manufacture self-driving electric vehicles. Now they are pushing these cars to market. Suddenly, they have become very good, just like Chinese smartphones and drones, they are fully modernized, very advanced, and very cheap, at the cutting edge of technology. The cars have also become very good, priced at only a third or a quarter of similar American cars.

The fourth stage is robotics. If you have the supply chains for phones, drones, and cars, you almost have everything needed to manufacture robots. This is the next stage. They are doing this. Of course, the U.S. has Elon and other companies making humanoid robots. I hope and expect they will do well. But China is definitely doing this too.

The company I am most focused on is a Chinese national champion called Unitree (宇树科技). We are not involved, but the robotic dogs that Unitree sells are comparable to Boston Dynamics' robotic dogs. Boston Dynamics' robotic dogs are priced between $50,000 and $100,000, which is why you rarely see them. Unitree's dogs start at $1,500, by the way.

We have two, and they are great. They can do backflips, they can climb stairs, they can talk to you, they are built with large language models, and they can teach you quantum physics while running around in your yard, which is fantastic. Then they are also starting to roll out humanoid robots now, at much lower prices. They are definitely moving in the direction of robotics.

This will be a real tug-of-war if you believe humanoid robots will emerge, and I do believe that, and on a large scale, if China is willing to manufacture them for $10,000 or $20,000, we could buy a billion of them, and suddenly we have robots building houses, doing gardening work, doing everything you want robots to do, waiting to serve you, then China manufactures them and sells them to you, and they are very cheap and work very well, which is fantastic.

……

Phones and drones are already a fierce issue, but cars and robots will be even more intense. This hasn’t fully happened yet because the robotics field hasn’t fully exploded, but I believe the robotics field will explode in the coming years.

Patrick: It’s fascinating to watch the race to create bodies and brains for robots. Companies like Physical Intelligence in the U.S. are working hard to build the datasets we don’t yet have, similar to the open networks we once had to train AI. Do you see some exciting areas where many young people and companies are making you feel excited, but you think the market hasn’t yet realized what’s happening and the potential?

Marc: I think it might be biotechnology (Biotech). The good news is that in the modern world, many people are interested in new technologies, and many people are talking about it. When I was a kid, the early adopter market was very small. So, the people who wanted their first personal computer or something were just a handful.

Now you have 50 million or 100 million early adopters who just want the latest thing and are constantly talking about it online. So, I’m not sure if there is too much delay now, but perhaps in the field of biotechnology, everything from life extension, embryo selection, possible reproductive technologies, to obtaining embryos from stem cells, for example.

Obtaining embryos from stem cells, you know, you might know many people who have such situations, where people have fertility issues when they are young, or they reach a certain age and have fertility issues, but they want more children, and then they are forced to make some tough choices involving IVF or different types of donors.

It seems we will be able to obtain embryos from stem cells, so you could have biologically meaningful children at a later age. External pregnancies are still a while away, but maybe at some point, this will be a big issue. People often talk about birth rates. Well, if you can continue to have children in your 60s, if you can have a dozen children through external pregnancies, would more people choose to do that? Maybe.

So that’s one aspect. Another might be gene optimization. So, an endless hot topic is intelligence enhancement. Now we have CRISPR, we have gene editing technology.

Then scientists are figuring out hundreds of genes that correspond to IQ. So, you should have the ability to enhance IQ, which raises a series of downstream questions.

……

Patrick: Very interesting.

Marc: Everything I just described is becoming possible. They have incredible implications for health, society, and so on, which will manifest over the next few hundred years. So, I think people may start to realize that there is more discussion to be had in these areas than what we are doing now.

……

Patrick: A quick final question. …, besides the previously mentioned "Machiavellians," which book would you choose?

Marc: I still very much resonate with a book called "The Weirdest People in the World" by Joseph Henrich. This book is probably about ten years old, but I think it hasn’t received much attention. This book is very insightful for understanding the nature of culture, especially the nature of different cultures.

As you know, there is so much in our current politics related to Western culture, and the implications of immigration, all these different debates, etc. For me, this is the most informative book trying to understand how to think about culture.

Patrick: Marc, thank you very much for taking the time.

Marc: Alright. Thank you, Patrick.

Original link: https://joincolossus.com/episode/the-battle-for-tech-supremacy/

免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。