jakiro
From August 17th to 23rd, I participated in the 2025 Silicon Valley Crypto Industry Study Tour event organized by Uwe.I have not participated in this form of study tour in full, this is the first time.In the past, when I visited the distance and studied, I felt prejudice in distance, and I felt quite disapproving.What can a group of people who were originally unfamiliar with their own different purposes and gathered together temporarily and took a quick look in a short time?I was very suspicious at first.Especially when I see some study tour summary written by others, which is full of advertising and excessive force, it is suspicious.
This time I personally participated in the whole process, and unexpectedly, it was very rewarding.Silicon Valley is the center of global technological innovation. I have many old friends in Silicon Valley and have visited them more often.Especially in the past year, I have been there several times for business and personal reasons, so I don’t have high expectations before this trip.The places that come and go often are also popular attractions for Chinese people to make pilgrimage to the United States. If you have read too much content that you sincerely share and show off, what else can you find?However, this time, under the leadership of Principal Yu Jianing, more than 20 members of the group spent a week of rich content and explosive information. To be honest, he was full of gain.Afterwards, I asked myself, why did I go there several times and meet many people, but I didn’t get such a big gain?The answer seems to be obvious. When you go to Silicon Valley, you can only use part of your resource network. The study tour group can aggregate and superimpose the resource networks of organizers and all group members, form an energy field, and release them in a concentrated manner, forming a lot of collisions with the ideas in your mind and triggering a lot of thinking.After returning to Australia, I have settled for a few days, and now I feel it is necessary to summarize some of the main gains of this study tour into text and share with readers.
Because of this study tour, my previous series of summary articles from the Hong Kong trip were also delayed.However, it was also because of the loss of his career. After a comparison, he became more aware of the situation in Hong Kong crypto.During this period, I will concentrate on writing down the summary articles of Silicon Valley and Hong Kong.But while I still remember it fresh, I want to join the team and send out the summary of the Silicon Valley tour first.Some readers have been urging me to update my Hong Kong bank summary in the background. Please wait, I will definitely not be able to do this time.
The content of this study tour is very rich and it is impossible to record it in a complete way. I just summarized the points that I had the deepest experience into words.Each section has a topic, and each section can be regarded as an independent article.It only represents my personal thoughts, and neither can fully and objectively reflect the entire picture of this study tour, nor does it constitute any investment advice.
A new round of Crypto prosperity is becoming a consensus
My main discovery of this trip to Silicon Valley is that the American technology community has generally formed a consensus that a round of crypto prosperity is coming.Everyone we meet and communicate, regardless of the degree of participation in crypto, whether it is cognitive about crypto, or even some people who are biased against crypto, think that crypto will usher in a prosperous period.When making this conclusion, they did not consider the attitudes of other countries and regions too much, such as whether China and Europe will follow up, but instead believed thatEven if other regions around the world do not follow up or respond, relying solely on the policies, funds and technical conditions of the United States is enough to create such an industry prosperity..
The root of this optimism is, without any hesitation, mainly the Trump administration’s new crypto policy.In addition to actively promoting the stablecoin Act (GENIUS Act) and the CLARITY Act, core members of Trump’s team publicly support crypto on various occasions and have been involved in the relevant business in depth.Therefore, with the gradual aggregation of various conditions, people in Silicon Valley generally believe thatThe crypto industry is about to usher in a long and large-scale prosperity, and the United States is not only the source of this round of crypto prosperity, but also the center..
This judgment conveys a sense of urgency.
On the last day of this trip, I visited the Huang Renxun Building of Stanford University’s School of Technology and saw the following exhibit: the first Google server hand-built in Stanford’s doctoral dormitory in 1996. The outer case was spliced with many Lego bricks.This is a famous exhibit at Stanford Institute of Technology, and I believe many people have seen it.But people of my age still have different feelings when they see it.Someone who once knew the history of China’s Internet told me that probably before and after this server was born, China’s management methods for the Internet were lucky enough to choose the right direction.At that time, there was a plan for the telecommunications department to manage the Internet in the way of managing landlines. If we really took this path, today’s Chinese Internet would probably be the same as everyone else, and the Chinese story of the past twenty years will definitely be beyond recognition.Fortunately, China made the right choice at that time, and the Internet industry has made great achievements in the past 30 years.Times have changed. I wonder if people in thirty years later can still talk about China’s blockchain industry with a kind of confidence that “we have not missed it”.
The “stablecoin war” did not come as expected
When the U.S. stablecoin bill was passed in July, a global wave of discussions on stablecoins began.At that time, I once had a judgment that once the bill was passed, all institutions and companies that had the conditions to issue US dollar stablecoins would quickly issue US dollar stablecoins, and a stablecoin war would occur in a short period of time.One of my purposes this time when I went to Silicon Valley was to find out if this was happening.
Why is this matter important?Because it is related to the popularity of “stablecoin payment”.Anyone who knows this field knows that stablecoins have a great improvement in cross-border payments compared to traditional payments.However, traditional payment is currently a field that is fully competed, and there are powerful players on all tracks.Although stablecoin payment is advanced, there is no powerful solution service provider yet, so it is not easy to pry a gap.What can truly promote stablecoins to quickly impact traditional payments is to catalyze existing payment companies and banks to actively issue stablecoins.A stablecoin war can create a sense of urgency and greatly accelerate the process.
But unfortunately, the Silicon Valley campaign has forced me to admit that my original view was wrong.We have not seen the “stablecoin war” happen.This is not to say that there are no new stablecoins in planning and design. As I was writing this article, Hyperliquid announced the issuance of a new US dollar stablecoin USDH, which of course proves that new players will continue to join the stablecoin team.However, there is indeed no “scrowding” situation as imagined, especially banks and Internet platforms. They were originally players who were able to get the best benefits from issuing stablecoins and most powerfully promote stablecoin payments to the real economy, but so far they have remained quite restrained and calm.
Why did the stablecoin war fail to come as scheduled?I have the following three guesses.
The first is that these banks and Internet companies are not ready yet.Trump’s election has caused the US stablecoin policy to turn around by 180 degrees in almost half a year, and these large institutions may not have time to make decisions.
The second reason is that the threshold set by the bill plays a role in blocking ropes..During the study tour, the organizer invited Coinbase’s senior legal counsel to communicate with us. The person mentioned that Coinbase has strongly influenced the strategy of the US crypto industry through very effective political donations, including legislation of bills such as GENIUS, so that some targeted blocking measures were carefully designed in the bill to accurately block potential opponents from entering the US dollar stablecoin market.This can be considered a strategy for self-protection of the “first-comer” in the crypto industry. Although it sounds unauthentic, it is actually reasonable.
The third reason, perhaps the deepest, is the “innovator dilemma” revealed by ChristensenThat is, because new technologies conflict with the current main business and existing interests pattern of enterprises, the innovation departments of large market-leading large enterprises are suppressed everywhere in corporate politics and cannot promote disruptive innovation internally, especially when such innovation is controversial.This effect has been discussed a lot, and there are also bloody examples like Kodak and Nokia, which are not far away, but it is actually difficult to overcome.
Stablecoin payment is a typical controversial disruptive technology.If it is a common word like AI, there will actually be no innovator dilemma.Afraid of blockchain is the most likely to cause large organizations to fall into the dilemma of innovators.Until now, many people still have doubts about the technical and economic value of blockchain payments, and even stubbornly deny them.They often say toughly that blockchain payment has no advantage over a proprietary technology they have been carefully crafted for a long time.Ordinary users cannot understand open systems and network effects, which are very “virtual” but powerful. With just a superficial user experience, they cannot tell which one is right or wrong, nor can they send clear signals to the decision-making level.The result will make the blockchain department within large enterprises extremely weak, and they are at a disadvantage in the fight for various internal resources, and it will be difficult to show any decent business achievements to convince the CEO.I have contacted many blockchain departments in banks and payment companies, and they are almost all struggling on the edge of the entire company, and it is difficult to cause meaningful changes within.To this day, many laymen still believe that the opportunity for stablecoin payment belongs to those large banks or Internet payment companies that already have a strong user base. I believe that the probability of exchanges and cross-border e-commerce companies winning in stablecoin payment is much higher than that of banks and Internet payment companies.
But now I have to admit that the expected stablecoin war did not happen immediately.But I still believe that such a situation will occur, but it is still brewing.
The Crypto industry enters RWA cycle and will soon be replaced on a large scale
One of my biggest questions before going to the United States wasIs there still a copycat season”. I hope to form a clear judgment through this exchange.
The so-called “counterfeit season” is the abbreviation of the “counterfeit bull market” in the industry.The so-called “alliance coins” originally refer to all digital currencies other than Bitcoin, but as digital currencies such as Ethereum have gained a foothold after experiencing the test of time, the current “alliance coins” are an asset category corresponding to the “mainstream coins”, which generally refers to those digital assets with small market value, low liquidity and low ranking.In the past two major bull markets, altcoins have collectively soared dozens or even hundreds of times, which is called the “counterfeit season”.
The usual pattern of Crypto bull market is that Bitcoin first realizes price recovery, then breaks through the previous high, causing a huge increase, and then Ethereum awakens to grow at a speed and amplitude that exceeds Bitcoin, which then drives the arrival of the copycat season.Usually the copycat season is the climax of the bull market, forming new assets for the industry, incubating a new generation of mainstream projects, and laying the root cause for market collapse.
Since this bull market, Bitcoin, Ethereum and other mainstream digital currencies have achieved breakthroughs one after another. The above model has gone halfway. The key question below is whether the copycat season will come as scheduled?
During this trip to Silicon Valley, we have come into contact with many crypto institutions and experts. Through communication with them, I have made a clear judgment that this bull market will not have a “copy season”.Or to put it another way, there will be a “copy season” in this bull market, but the chips in it are different from the previous few times. They are no longer dazzling and strange crypto altcoins, but RWA concept assets, so they can no longer be called “copy season”.
I have three reasons for this judgment.
First,The dominant player has changed.The last bull market occurred against the background of unprecedented currency releases in the world and the direct cash from many governments to families. Retail investors have become unprecedentedly strong and play a major role in the bull market.They not only set off a bull market in the currency market, but also competed with mainstream institutions on Wall Street.But starting from 2022, the currency has entered a tightening cycle, and the market collapse has eliminated most of the wealth in retail investors, and only institutional investors still have funds in their hands.So from 2023, the situation has become very obvious, with Wall Street and institutional investors becoming the main players in the market.After the US crypto policy shifted, a large number of professional institutions entered, further consolidating the dominance of institutions in the crypto market.The addition of these professional institutions will lead to fundamental changes in the liquidity preferences and compliance awareness of the entire market.I doubt whether the counterfeit projects in the past that were carrying a few demos and telling big stories could be recognized by the leading institutions.
second,The market mentality has changed.With the addition of mainstream institutions and mainstream funds, the mentality of entrepreneurs and investors has also changed.Some Silicon Valley Crypto VCs I was familiar with in the past have adjusted their aesthetic views, paid more attention to projects related to stablecoins and RWA, and paid more attention to equity investment.The rash issuance of altcoins has been considered a negative indicator.
third,The industry theme has changed.This bull market is undoubtedly based on RWA.Please note that in the Chinese community, when RWA is mentioned, many people quickly think of very “real” and “tangible” assets such as houses, forests, land and mines, jade and antiques.But in fact, bonds, equity, copyright, securities, etc., are the “virtual assets” in the real world that are larger and easier to operate.From the perspective of RWA implementation logic, it is necessary to first put these virtualized, homogenized, and securitized RWAs first on the chain, and then gradually it will be those very “real” assets that are visible and tangible.Currently in the United States, the current hot topics of RWA are particularly specific and focused, which are the US stocks.If you expand a little bit, high-quality unlisted companies’ equity will also be in the focus.
This time we realized one thing in Silicon Valley, that is, if all types of corporate equity can issue tokens, the competitiveness of altcoins in the pure currency circle is very weak. Most of them are based entirely on the transaction and speculation of digital assets. They have nothing to do with the real world or have extremely weak relationships, and the team does not have the resources and experience in the real world.In comparison, in the traditional market, there are a large number of high-quality companies’ stocks, equity and other interests that have not been tokenized yet.If those high-quality project equity of AI, biomedicine, new energy, and smart hardware issue tokens and enter the crypto market, can those altcoins that “hide into small buildings and become unified” compete with them?
Of course, I also know that there are some “bankers” in the crypto circle who are well versed in human nature and have rich experience. They are very good at playing games of building momentum in bull markets. In every bull market, they will make some achievements.But overall, I think with the influx of quality RWA assets, the theme of the market will be switched to RWA assets, after all, there is only so much liquidity.
At present, the RWA is just brewing, after all, the CLARITY bill has not been passed.However, institutions interested in crypto have begun to adjust their aesthetic views in advance, and I believe that the market will soon change funds on a large scale.For counterfeit projects, the opportunity for RWA must be grasped in some way, otherwise the future situation will be even more unoptimistic.
From the perspective of crypto industry development, moving towards RWA is very beneficial because it marks the return of the crypto industry to an open system.
Looking back at the development of crypto, from 2009 to 2017, although the crypto industry was in infancy, the blockchain infrastructure was very primitive, and many scammers were pouring in, crypto was an open system at that time, and people were thinking about how to use new technologies to transform the world.But after 2018, as major countries adopted negative policies on crypto, crypto gradually deviated from its connection with the real world and became a closed system. Most crypto projects revolved around the “demand” of speculative gambling, causing the entire industry to become more and more vulnerable.This is also the inevitable fate of a closed system: after stopping the exchange of energy with the outside world, the entropy gradually increases, and eventually falls into “hot silence”, without any meaningful physical laws.When the entire crypto market was crazy at the end of 2024 and early 2025, there was no pattern on the surface and it was completely trapped in the big gamble of robots between dealers and retail investors. This is a typical closed system dead end.Fortunately, the crypto industry has not been hovering on this line of thinking for too long. With the launch of stablecoins and RWA, the entire system has reopened and began to exchange energy with the real world.Many people may not realize that this is actually the road to rebirth of crypto.
Three hot spots in Crypto in the United States
Summarize our exchanges with the crypto industry in Silicon Valley. Currently, there are three hot topics in the crypto industry in the United States, namelyCoin-stock linkage, US stocks and all things exchange.
The so-calledCoin stock linkageIt is to form a resonant relationship between the stock market and the currency market in some way.At present, the linkage of currency and stocks has substantial effects in the US and Hong Kong stocks, but the development stages of the two markets are different.The currency-stock linkage in Hong Kong stocks is still in the stage of speculation on blockchain and digital asset themes, while in the United States, the current main form is Digital Asset Treasury (DAT) companies.This time I have come into contact with some professional institutions in the United States, and they are basically actively exploring and operating DAT listing.This model has become the most mature and effective path on the market due to the successful practice of micro-strategy.However, from the perspective of currency-stock linkage, the degree is relatively shallow, not penetrated into the business level, nor does it mobilize the power of the token economy, and can only be considered “entry level”.The rapid pullback in the market in recent times has caused some concerns about the sustainability of this model.Some institutions with currency experience in Silicon Valley have begun to plan the DAT 2.0 model, but their own interpretations are on what this 2.0 is.Which mode can be used to run out, and it still has to be put on the market to race horses.I firmly believe that if the development of the DAT model continues, it will inevitably lead to the deep coupling of the token economy and business.
But even a simple DAT still has many details to operate.This time, we learned about the practical path of DAT from some institutions with practical experience in Silicon Valley, such as SPAC mergers and acquisitions and RTO (should be bought and listed), and felt that it was quite difficult and expensive.I think there should be more ideas about the currency-stock linkage, and it will not only be on DAT.
US stocks are listedIt is currently clearly in the process of brewing.Coinbase, Robinhood and Kraken all announced clear plans.Among them, Robinhood is moving faster and has launched US stock tokens based on Arbitrum in Europe, which will launch more than 200 US stocks and ETF products on the chain for sale.Kraken launches xStocks products for non-US customers, achieving uninterrupted trading on the Solana main chain, and currently supports more than 50 US stocks and ETFs.Coinbase is deeply involved in the U.S. market, viewing stock coins as part of its grand “Equity Exchange” strategy, actively applying for compliance paths from the SEC, including seeking no-opposition letters or law enforcement exemptions to achieve legal tokenized stock trading.
I originally believed that after the stablecoin bill was passed, the entire industry would spend some time digesting the achievements of stablecoin and fully promoting the implementation of stablecoin payments.But now it seems that my judgment is wrong. The entire industry does not love stablecoins, but is directly promoted to the US stock market.Promoting the implementation of stablecoin payments will affect the cheese of banks and traditional payment companies, and it involves all aspects, so I can only slowly take it.The forefront of the entire industry will undoubtedly move along the most labor-saving route, which is to quickly pull more high-quality assets onto the chain and form a trading pair with stablecoins.US stocks are undoubtedly the hot spot at the moment.I believe that in a few months, stocks will become the hottest topic in the industry.
The third hot topic is called“Exchange of Everything”.This time, I carefully learned about the direction of venture capital institutions’ attention in Silicon Valley and found that the exchange is the highlight of the focus.This is not difficult to understand.The exchange is the leader in the entire crypto industry ecosystem and the pinnacle of the food chain, but the competition in this track is very cruel, and it is difficult for latecomers to make a comeback.However, whenever the market changes chips and new product categories appear, they will bring new game rules, new user groups and new market structures, and will bring opportunities for reshuffle to the exchange track.Bitcoin in 2011, altcoins represented by Litecoin in 2013, and the explosion of Ethereum ERC-20 tokens in 2017 have all given birth to exchange giants that dominate for the time being.If this rule continues to be effective, then when the entire industry exchanges RWA, a new generation of exchanges will inevitably appear.
So what is this new generation of exchange like?In several speeches on crypto by Atkins, the new SEC chairman of the United States, he mentioned the so-called “super applications”, which are super platforms that can trade all-in-one products in one application.When this super platform was later digested by the US crypto industry, it was renamed “Exchange for Everything”, and now it has become an absolute hot spot for US crypto venture capital.
Currently, all things exchange all two assault paths.One is a centralized exchange of things represented by Coinbase’s “Project Diamond”, and the other is a decentralized exchange of things represented by Hyperliquid.Both have proposed a clear plan to aggregate all trading products such as digital assets, stocks, bonds, gold, foreign exchange, etc. on a single platform, and even Polymarket predicts the market to let go of these kilograms of meat and bones. It is ambitious and ambitious.
Whether we appreciate this gluttonous monster or not, the exchange of all things conforms to market laws and also conforms to the principles of network effects.Aggregating all trading products, all liquidity, all information and all users on a single platform is undoubtedly the holy grail in the trading market and the most efficient market.In reality, this kind of exchange is certainly impossible, but the process of advancing to this goal will undoubtedly produce a giant-level platform and will also profoundly transform the structure and game rules of this industry.
I hope the Chinese crypto industry can be prepared and made arrangements for this trend.
Silicon Valley relies on the “Acquaintance Circle” to form a unique innovation zone
During this study tour, the organizer fully considered the characteristics and richness of Silicon Valley. He did not put all resources on crypto, but arranged a lot of exchanges related to AI and other innovative technologies.In the first part of the study tour, a famous technical expert and science and technology writer Wu Jun was invited to give a comprehensive introduction to the basic situation of Silicon Valley and the global technological innovation trends.Then several heavyweight guests were arranged to communicate with us to help us understand the situation in Silicon Valley.
I have been to Silicon Valley many times and am familiar with some of the data in Silicon Valley.For example, Silicon Valley, including San Francisco, has a total area of 4,800 square kilometers, of which the core built-up area is only more than 500 square kilometers.The total population of the Bay Area is about 9 million, but the population of Silicon Valley technology is only 3 million.Teacher Wu Jun provided some more detailed statistics, such as the total number of employees is 1.7 million, software engineers is 150,000, and the annual household income is US$200,000, and the first generation of immigrants accounted for 40%, of which Chinese and Indians accounted for 6%. These are very interesting data.We always say that Singapore is a small country, but in terms of built-up area and population, Silicon Valley is still smaller than Singapore, but it has achieved such brilliant achievements.I believe that most people who visit Silicon Valley have a question mark in their minds, which is to understand “why this place is so magical.”
Silicon Valley is undoubtedly the most successful and most friendly technology innovation zone on the planet.From the perspective of venture capital, Silicon Valley and the San Francisco Bay Area attracted a total of US$69.7 billion in venture capital in 2024, an increase of 125% year-on-year, accounting for 52% of the total U.S. venture capital investment.In addition to those famous listed technology giants, Silicon Valley owns nearly 300 unicorn companies, accounting for nearly 40% of the United States.In a private exchange, a Chinese VC manager from Silicon Valley proudly told us that 20% of the world’s successful venture capital opportunities are within our 40-minute drive.A friend of mine’s Wall Street investment banker told me that Wall Street doesn’t look at startups unless it comes from Silicon Valley or Israel.
What exactly makes Silicon Valley have such strong innovation capabilities?
There are a lot of discussions on this issue, and even in Chinese literature, there are a variety of views. Some emphasize the role of Stanford, some attribute to the shaping of Silicon Valley culture by early technology companies such as HP, Fairchild, Intel and Apple, some attribute to Silicon Valley’s rich venture capital and industrial clusters, and more people simply explain that the smartest minds in the world have emerged.
But in fact, the above statements are not very convincing to me, because I think many of these explanations are caused by inverse results.Take “smart mind” as an example. In fact, it is not that the mutated microorganisms contained in the soil of Silicon Valley that can cultivate a large number of geniuses, but that geniuses from all over the world are constantly gathering in Silicon Valley.To be honest, the original place of technological innovations that have been pressed on Silicon Valley is not in Silicon Valley, nor is it in the United States. Instead, it later went to Silicon Valley to achieve a transformation from innovation to entrepreneurship.As a local Silicon Valley investor said during this study tour, the most awesome thing in Silicon Valley is not technological innovation, but assembling innovation, talents, funds and systems into a successful startup company.So my question is, what factors make Silicon Valley particularly good at incubating and cultivating startups?
The question I raised is largely due to my entrepreneurial experience in recent years.The practice over the past few years has made me deeply feelStarting a business cannot be separated from venture capital. Venture capital is based on integrity, and integrity is the most difficult quality in investment..A startup project can win the trust of investors, get hot cash, and then in the rules of the game where you can get rid of it even if you lose all your money or put it in your own pocket, you can also overcome all kinds of difficulties and obstacles, resist all kinds of temptations, and do the project in a solid manner. This is really a very contrary to human nature and requires huge willpower and self-PUA to do it.On the one hand, without sufficient financial support, most startups will not be able to reach that turning point.On the other hand, for investment institutions, it is indeed a super difficult thing to identify honest and capable entrepreneurs.Therefore, in China’s venture capital industry, terms such as “bet” and “repurchase” are often introduced to protect investors’ rights and interests.But this approach completely passes on entrepreneurial risks to entrepreneurs, breaks away from the essence of venture capital and can only curb innovation.I have complained about these deteriorating “VCs” in China on many occasions before, thinking that they are China’s “features”.But after going abroad, I found that in fact, most so-called “VCs” in Asia come with one or another toxic clauses, whether in Hong Kong, Singapore or Malaysia.In comparison, the authentic venture capital in Silicon Valley is the real venture capital and is a special case.This made me more and more curious about how Silicon Valley balances trust and constraints and becomes a unique Feng Shui treasure land for entrepreneurial incubation in the world?
This trip to Silicon Valley has given me some new views on this issue.Through my limited contact with the Silicon Valley venture capital circle, I seem to feel thatVenture capital in Silicon Valley is actually based on the “circle of acquaintances”.Entrepreneurs and investors often use classmates, colleagues, and likes as coupling points, and use one strong relationship after another, and connect, select and eliminate them for a long time with very high standards, convey trust, provide commitments, and impose constraints to each other, form a clear-level trust circle.As for investors, entrepreneurs who can be successfully selected into these circles have undergone screening and long-term inspection, and are subject to soft constraints in many aspects and are trustworthy.As for entrepreneurs, once they reach this point, they can gain trust and resource support that is difficult to obtain in other places. They need money to pay, people to give people, and relationships to give relationships. The probability of success will naturally increase by a hundred times. The consequences of “making trouble” are extremely serious. With rational choices, they will naturally follow the good.
In other words,Although Silicon Valley’s products are the world’s top technology innovation companies, the mechanisms they rely on are very ancient acquaintance society and circle culture, rather than relying on advanced game mechanisms or innovative financial tools.It is precisely the compression of the two groups of entrepreneurs and investors into a small circle and all-weather contact that can form this innovative incubation mechanism with strong binding and support, guiding and forcing entrepreneurs to follow the right path, and providing the greatest resource pressure to help them succeed.Many people feel magical because of the “small” of Silicon Valley, but in fact, Silicon Valley’s success is precisely because it is small enough that this mechanism of acquaintance circles can work.
I communicated this “acquaintance circle theory” with several friends in Silicon Valley and gained their recognition.My question is, can this experience be copied?
How does Silicon Valley view the “AI bubble theory”?
When you come to Silicon Valley, you can’t just talk about crypto.When you come to Silicon Valley, you can’t avoid AI.In fact, the center of crypto in the United States is in New York, and the absolute theme of Silicon Valley is AI.Before this study tour, I had already made up my mind to understand what Silicon Valley thinks about the “AI bubble theory” that is now growing wildly.
Speaking of the “AI bubble theory”, many people may still be unfamiliar with it.As far as I read on the Chinese Internet, now I am basically bullish on AI, which has almost evolved into a kind of political correctness.But in fact, the concerns about the AI bubble have actually become increasingly amplified over the past few years.Since ChatGPT 3.5 made a breakthrough, a group of experts led by Meta’s chief AI scientist Yann LeCun have been publicly skeptical and even critical of whether the Big Language Model (LLM) can achieve strong artificial intelligence (AGI).With the release of ChatGPT 5, the development of core capabilities of large models has slowed down significantly. Under this circumstance, the “AI bubble theory” has begun to ferment recently. Its representative is Gary Marcus, an American cognitive scientist and professor of psychology at New York University.
Gary Marcus is a longtime critic of Silicon Valley and the AI connectivist route.His criticism of Silicon Valley and artificial intelligence mainly focuses on three aspects.First, he believes that Silicon Valley is overly pursuing short-term business interests and touting artificial intelligence as a “universal antidote”, but ignores the true limitations of technology and social risks.Secondly, he criticized the technical path of large models for relying too much on massive data and computing power, and satirized that deep learning is “automatic completion of drug use”, lacking in-depth research on common sense reasoning, causal understanding and transparency, which leads to AI often showing false intelligence in real-life applications of “willing to say but not understand”.Again, he pointed out that Silicon Valley companies lack sufficient ethical responsibility, are keen on hype and narrative packaging in the capital market, but are unwilling to face the security, bias and regulatory issues that AI may bring.These are Gary Marcus’ long-term views, but after the release of ChatGPT 5, he became particularly active, actively receiving various media visits, claiming that the AI potential of this wave of big models is about to run out and the bubble is about to collapse.
So how do people in Silicon Valley view the “AI bubble theory”?
I encountered three different attitudes during my study tour in Silicon Valley.
The first attitude is to sneer at this, believing that the AI bubble does not exist at all and the prospects are bright.Some investors have strong confidence in the development prospects of AI. Although they do not deny that ChatGPT 5 is lower than expected, they remind us to note that the revenue of AI leading companies is growing rapidly and their valuations are soaring. There may be unlisted AI companies with valuations of 500 billion or even higher next year.
The second attitude is to be very firm in believing that AI is bubbled and very large.A well-known scholar who was inconvenient to mention his name replied when I asked about this that the bubble level of AI companies in Silicon Valley today is no lower than that of China’s “Four AI Dragons” back then. Although I don’t know when the bubble will burst, once it is broken, Silicon Valley will suffer a heavy blow.Another new Silicon Valley investor predicted the bursting time of the AI bubble and has begun to make corresponding arrangements for investment, preparing to turn bad things into good things.
The third attitude is to believe that although AI has bubbles, bubbles are not a bad thing, and AI will “dance with bubbles” and continue to move forward.This view is relatively mild and compromised, and represents the opinions of most people.
In principle, I support the third view, but I do have my own concerns about the current AI development model.
The biggest feature of AI with deep learning as its core is its opacity, which is the first “unexplainable” black box in the history of human science and technology development.The most advanced AI model now has hundreds of layers of artificial neural networks, which exceeds anyone’s explanation ability.So neither do anyone knows why it is so smart, nor do anyone knows why it isn’t smart enough.One direct consequence of “unexplained” is that when AI’s capabilities are less than expected, people have no better way than to constantly accumulate computing power.Each time you add a little computing power, the effect will be better.If the effect is not good enough, then increase the computing power.No one knows where the end of this technical route is, whether it can really lead to AGI before exhausting the energy available.We can hear various optimistic or pessimistic views on the dinner tables of self-media and Silicon Valley, but these views are based on personal beliefs. No one can penetrate these hundreds of layers of neural networks to find out the truth.
The United States announced that the growth rate of fixed asset investment in August was as high as 5.4%, surpassing China for the first time in more than 30 years.However, nearly half of these fixed asset investments are directly or indirectly related to AI infrastructure.A large amount of resources are used to build extremely large-scale computing centers, and the available data on the entire Internet has long been exhausted.The current model is that the entire industry is committed to building a larger temple, hoping to summon the gods of AGI. This model is a huge test for people’s patience.
Is Silicon Valley worried about this?I think the disagreement has emerged.If the core capabilities of the big model cannot restore the momentum of rapid improvement, and if the intelligent progress that allows the public to have a clear perception cannot be quickly generated, but only continues to expand applications at the current level, then in the near future, Wall Street and the American public will be angry at the endless investment in computing power at a scale of hundreds of billions of dollars.
Silicon Valley also has limitations
When the study tour was about to end, I and several friends who were traveling with me summarized the gains from this trip. We all felt that Silicon Valley also has its own closure and limitations.Silicon Valley is indeed passionate about technological innovation and full of piety from technology pilgrims, but the other side of this passion and piety is its indifference to many major issues in the outside world, especially to ordinary people.
Recently, at the Technology Leaders Summit hosted by Trump, Silicon Valley technology leaders led by Zuckerberg and Cook have successively launched technology investment plans with a total of more than one trillion US dollars, which has made Trump happy.However, some American media pointed out that most of these large plans that cost hundreds of billions of dollars are invested in data centers, AI chips, supporting energy and other fields. They are huge in scale and advanced in technology, but they have little effect on improving US employment.
Does Silicon Valley care?Actually, I don’t care.Silicon Valley has its own concerns, and it almost only cares about these issues.Regarding the external world, they have a self-sufficiency and a indifference that does not matter to themselves.Startups in Silicon Valley only raise money from Silicon Valley, and the huge increase in wealth created before listing is almost exclusively distributed within Silicon Valley.People in Silicon Valley do not entertain, eat just the same food, and receive nominally high salary, and most people’s lives are quite boring.But these food, drink and fun are not important in Silicon Valley.What is really important is whether you can gain an advantage in the technical competition, whether you can get the next round of financing, whether the project valuation curve is upward, whether the algorithm performance is improved, whether the infrastructure is expanded, and whether you can prove that you are a real talent in the fierce competition.As for whether these efforts can truly solve problems such as education, medical care, the gap between the rich and the poor in the outside world, or cause the worsening of these problems, they rarely care.For Silicon Valley, the outside world is an abstract object that needs to be transformed and optimized by their technology, rather than a community of hundreds of millions of people like them that need to be shared and responsible.
A very popular idea in Silicon Valley in recent years is Dark Enlightment represented by programmer and well-known blogger Curtis Yarvin, which is also summarized as “neo-reactionaryism” and is classified as a branch of neo-fascism in the political genealogy.This faction opposes democracy and supports Trump, comparing the country to a joint-stock company, and managers selected by “shareholders” are effectively governed rather than relying on referendums to compete with political parties.Because Curtis Yarvin uses the pseudonym “Mencius Monbag” as his pseudonym, many people in the Chinese circle call him “Mencius Yawen”.
Mencius Yawen’s thoughts were fueled by Silicon Valley giants such as Peter Thiel, and were accelerated in Trump’s victory in the election, subtly affecting many people inside and outside Silicon Valley.Although some of the people we contact in Silicon Valley may not have read Mencius Yawen’s articles, or even don’t know this person at all, many of the thoughts in their minds are shaped by Mencius Yawen.For example, technology first, efficiency first, and elite governance.They are not concerned about the issues that are plaguing the American people and the world as a whole, but are more focused on how to build a “small universe of technological innovation” in the Bay Area.Their ideal order often emphasizes the replacement of political consultation by technological means, capital and computing power as sources of legitimacy, public responsibility is simplified to return on investment, and society is conceived as a company that needs management and optimization.Some people summarize this arrogant worldview in Silicon Valley as “technical aristocratic.”
I personally think that in a small area on the earth, it is the pride of human civilization that has long existed such a pretentious and infinite creativity.There is no need to ask people in every part of the world to understand the righteousness so deeply.As long as they can continue to innovate, make them arrogant.However, we should also realize that not all questions can be answered in Silicon Valley, and Silicon Valley does not care about many things.
There is a view that Silicon Valley was not in the past and will not become a crypto center in the future, because crypto focuses on fairness, while Silicon Valley focuses on efficiency. The basic spirit of crypto is to be equal to all beings in the face of cryptography and smart contracts, which runs contrary to Silicon Valley’s technocracy.
I’m not sure if this view is correct.For the crypto industry, we should not hope to find all the answers in one place.But, I believe in Silicon Valley’s inclusion.Although Silicon Valley is not big, it should be able to accommodate AI and crypto.
