By Sleepy

Yesterday, I saw an article with the following headline: "Global AI New King Born! Anthropic Valuation Skyrockets to $1.2 Trillion, Overtaking OpenAI for the First Time."

This headline truly captures the essence of the current era - featuring AI, a comeback story, the ascension of a new king, and a staggering number that exceeds people's imaginations. It's like a gong. Once the gong sounds, it's hard for us not to look up.

So how did this $1.2 trillion valuation come about? It actually originated from the on-chain Pre-IPO market.

The so-called on-chain Pre-IPO market doesn't involve trading the ordinary shares you see in your brokerage account. It's more like a designed "pre-listing risk exposure" for a company. Through tokenization, SPVs, or synthetic structures, individuals split the future listing expectations of a private company into smaller units and facilitate trades on-chain. This opens up a window to ordinary investors that was previously hard to access and provides the market with real-time pricing. Anthropic was valued at $1.2 trillion in this market.

Over the past two years, the feeling AI has left with the general public is often not "I'm participating in a new era" but rather "the new era has passed me by." NVIDIA's stock soared, cloud providers surged, big AI model companies went through round after round of funding, yet the true core equity remains locked in private markets. We can see the ship but can't get a ticket. Therefore, any ticket that could potentially lead to companies like OpenAI and Anthropic comes with its own filter.

However, in moments like this, it is even more crucial to extract the numbers from the headlines and lay them out on the table to understand how they were derived. Anthropic may be one of the most worthy AI companies to be seriously studied at present. But the issue lies in the fact that a good company, a grand era, and an aggressive price tag do not automatically merge into one.

On the Jupiter crypto trading platform, Anthropic's Pre-IPO token had a daily trading volume of only $1.39 million, with just 329 traders in the past 24 hours. And yet, it is this $1.39 million and 329 traders that project the illusion of a trillion-dollar valuation.

However, I don't want to discuss whether Anthropic is truly worth its value or delve into the issues surrounding on-chain trading of Pre-IPO assets. I first want to clarify a more fundamental question: what criteria must a price meet to qualify as a "valuation"?

Price's Birth Certificate

In February 2026, Anthropic completed its Series G funding round. The company raised $300 billion at a valuation of $3.8 trillion, with lead investors being the Singapore sovereign wealth fund GIC and hedge fund Coatue Management. One month later, OpenAI also announced the completion of its latest funding round, raising $1.22 trillion at a valuation of $8.52 trillion, with major investors including SoftBank, Microsoft, and other institutional investors.

How were these figures generated?

Take Anthropic's Series G funding as an example. GIC manages over $700 billion in sovereign wealth funds, while Coatue manages a global tech hedge fund of over $600 billion. Each has teams of dozens of analysts who spent months analyzing Anthropic's technical architecture, revenue trajectory, customer retention rate, and competitive landscape. The final $300 billion investment came with full legal terms, including anti-dilution protection, priority liquidation rights, information rights, and board observer seats. If Anthropic underperforms or goes downhill, these terms ensure that GIC and Coatue can recoup their investment first.

What they purchased was not just a number but a complete set of legally enforceable rights.

What about the $1.2 trillion on Jupiter? Over three hundred traders, over one million dollars in daily trading volume, with no commitments or obligations from Anthropic behind the token. What you're buying is not a small piece of ownership in the company but just an on-chain bet slip.

Both prices are presented in headlines in the exact same way, referred to as 'valuation of XX billion'.

In 1985, financial economist Albert Kyle published the classic paper "Continuous Auctions and Insider Trading," introducing the concept of 'market depth' and using λ to measure the impact of a unit fund inflow on prices. In a deep market, a $100 million buy order may only cause a 0.1% price fluctuation, whereas in a shallow market, $50,000 could result in a 20% price swing. The larger λ, the greater the price impact of a single trade, and the scarcer the consensus information carried by the price itself.

In Anthropic's case on Jupiter, with a $1 million liquidity pool depth supporting a $1.2 trillion implied valuation, the liquidity-to-valuation ratio is approximately 1:1,200,000. If someone tried to sell a $10 million position at a $1.2 trillion valuation in this market, the entire liquidity pool would be drained ten times over. This price is untradeable; it only exists on paper and cannot be realized in the real world.

If it had only been treated as a reference indicator, that would have been understandable. The issue is that it was not treated that way. It became the argument for "officially surpassing OpenAI," the headline for the "birth of a new global king," and the cognitive input for many readers.

This packaging of the thin market's marginal price as broad consensus is not a new phenomenon. It has been happening for almost four centuries.

Tulip Mania

February 3, 1637, in Haarlem, the Netherlands.

In a small tavern, about thirty people sat around a long table. According to the customs of Amsterdam and Haarlem at the time, these informal tulip bidding gatherings were held several times a week, usually in the back room of a tavern. The participants were traders and florists who were familiar with each other.

The item up for auction that day was a Semper Augustus bulb. Its red and white petals were considered a masterpiece of nature, with only about a dozen bulbs known to exist in the entire Netherlands. The bidding lasted all night, and the final price was 10,000 guilders.

In Amsterdam in 1637, a canal-side townhouse was priced at about 5,000 guilders, and a skilled craftsman's annual income was around 300 guilders. One bulb was equivalent to two townhouses or 33 years of a craftsman's wages.

And the birth of this price came solely from thirty people, a confined space, alcohol-fueled energy, without any external constraints, market maker obligations, or public information requirements. Bidders emotionally inflated each other, and apart from payment, they had no obligations for their bids.

The next day, the transaction price was recorded in a booklet printed in Haarlem. This booklet was then carried by post to cities like Leiden, Rotterdam, and Utrecht. Farmers and small shop owners who read it had no way of knowing how this number was determined. To them, the price in the printed booklet was the market price. Some people started hoarding common variety bulbs based on this information, believing that the entire market would rise.

On February 6, at a tulip auction in Alkmaar, suddenly no one bid. This was followed by Haarlem and Amsterdam. Within a day, buying interest disappeared across the Netherlands. Those who had hoarded bulbs at the market price found no takers, leading to a price crash of over 90% within a week.

In hindsight, the "10,000 guilders" for that Semper Augustus was not a market judgment but a room's judgment. However, through the amplification of printing, the room's judgment turned into national cognition.

Eighty-three years later, 1720, London.

The South Sea Company's stock rose from £128 at the beginning of the year to £1,050 in June. This trading company, established in 1711, held a monopoly charter for British trade with South America, but the actual trade profits were extremely thin. The true driver behind the surge in stock price was a complex debt-for-equity swap scheme, where the company proposed to take on national debt and convert it into company stock, then sustain the entire cycle by continuously driving up the stock price.

Newton sold his South Sea Company stock at £300, making a profit of £7,000. However, the price continued to soar. In July, Newton bought back in, this time buying at a high point. After the autumn crash that year, his total loss amounted to £20,000, roughly equivalent to his ten-year salary as Master of the Royal Mint.

Newton perhaps never pondered how that "£1,050" he referenced came to be.

In 1720, there was no electronic trading system, no central counterparty clearing. Trading South Sea Company stock required a personal visit to the company's office in London for transfer, or through brokers in a few coffeehouses on Exchange Alley. Daily actual transactions may have been only tens to hundreds, with the direct parties involved amounting to just over a hundred people.

These prices were recorded on a price list at Jonathan's Coffee House. When newspapers reprinted these price lists, they did not include notes like "Today's 12 transactions totaling around £8,000." All readers across England saw was the single number "South Sea Company: £1,050."

When the panic selling started at the end of July, the price generated by that limited game among hundreds was instantly broken through. There were no market makers obliged to buy, no circuit breakers, no central bank intervention. By December, the stock price had fallen back to £124, almost returning to the starting point of the year.

Fast forward two hundred and sixty years. Late 1980s, Tokyo.

"The land under the Imperial Palace in Japan is worth more than the entire state of California." This statement was widely quoted by the global media in 1989. According to estimates at the time, the 2.3 square kilometers of land where the palace sits was valued at around $850 billion based on surrounding land prices, while the assessed total value of all land in California was about $500 billion. However, this estimate only considered the prices of a few actual transaction plots in the Ginza and Marunouchi areas.

Japan's land market has a unique structural feature, with an extremely low turnover rate. Japanese landowners often see real estate as a family asset to be passed down through generations, rather than for trading arbitrage. At the peak of the bubble in 1989, the total number of annual land transactions in Tokyo's core commercial district was extremely limited. Occasionally, land parcels that entered the market did so due to owner bankruptcy or family inheritance disputes, with a large group of cash-rich and eager buyers vying for the scarce supply.

The price arising from this extreme supply-demand imbalance has been extrapolated by real estate appraisal agencies as the "fair value" of all land in the area. Their logic is that if this small piece of land is worth 20 million yen per square meter, then every adjacent piece of land should also be worth this price.

In 1990, the Bank of Japan raised interest rates consecutively, and banks tightened loan standards. When businesses were forced to sell real estate to repay loans, a true test of liquidity began. Sell orders flooded in, but there were few buyers. The actual clearing price was 50% to 80% lower than the so-called market valuation.

Subsequently, Japan's national land price index continued to decline for a full 26 years until it saw a modest recovery in 2016.

The taverns of Harlem, the coffeehouses of London, the real estate appraisal firms of Tokyo, the Jupiter DEX on Solana. Four scenarios spanning nearly four hundred years, sharing the same narrative structure:

A tiny number of participants generate an extreme price in an illiquid market → The media propagates it as a consensus → A larger audience makes decisions based on this → When true liquidity is tested, the price reverts.

The media evolves, from pamphlets, newspapers, telegrams, TV, to WeChat public accounts, but that core flaw has never been fixed. When the price is disseminated, its birth conditions are systematically omitted.

Why?

A Complex Story, Always Compressed into an Easily Spreadable Number

Business reporting faces a natural challenge: the real world is too complex, and the window for communication is too brief.

What exactly happened with a company often involves discussing funding structure, product developments, revenue quality, competitive landscape, equity rights, exit pathways, and market sentiment. Yet a headline is only one line, and a reader's attention span is only a few seconds. Hence, expressions like "valuation exceeds $100 billion," "market cap evaporates by $1 trillion," "unicorn born," "super app rises," become easily chosen compressions. It is the narrow gate that complex business information has to pass through when entering public opinion.

Of course, writers are also within this gate. We all know that explaining the birth conditions of a valuation is much more difficult than writing a impactful headline. The former requires patience, length, and the reader's willingness to stay; the latter only needs a sufficiently bright number to immediately convey "something significant has happened here." A headline written as "Anthropic On-Chain Pre-IPO Synthetic Asset Reaching Implied Valuation of $1.2 Trillion in a Low-Volume Market Extrapolation" may be more accurate, but it might also lose its dissemination power before reaching the audience.

If it were written as "Global AI New King Born," the situation would be different. It would be dramatic, with winners and losers, a throne, and a power shift that humans always love to watch. Communication is not a transporter of facts but more like a juice extractor. Facts go in, and what comes out is emotion.

The second reason is market structure. The Chinese business information environment lacks a crucial player: the short seller.

In the U.S. capital markets, a price detached from fundamentals will not stay safe for long. Short-seller research firms like Muddy Waters, Citron Research, and Hindenburg Research have business models centered around identifying targets whose prices far exceed levels sustainable by liquidity or fundamentals. They then publicly release reports while profiting from short positions.

They have a strong economic incentive to show the public the true origins of a number. Muddy Waters' 2020 report on Luckin Coffee was 89 pages long, involved 92 full-time and 1,418 part-time investigators, recorded over 11,000 hours of store footage in 981 stores nationwide, and cross-checked 25,843 receipts. All of this was just to prove one thing: Luckin's reported daily sales per store were fake, with the real number being roughly half of the claimed figure.

This level of adversarial research requires two prerequisites. First, there must be a short-selling mechanism to profit from "price regression." Second, there must be legal protection to prevent the suppression of short reports. Both exist in the U.S. stock market. In the Chinese A-share market and primary market, both are essentially absent.

The result is that no one can make money by questioning valuations, so no one has the incentive to ask under what conditions this price was generated.

Short sellers are not destroyers. They are correction devices in the pricing system. The result of removing the correction device is that price deviations can continue to expand without any resistance until one day they collapse under their weight. And every day before the collapse, the market appears normal.

The consequences of the combination of these two forces are not uncommon in Chinese business history.

In June 2015, LeTV's stock price on the Shenzhen ChiNext board reached its peak, with a market cap of about 170 billion RMB. Jia Yueting's depiction of LeTV's ecosystem spanning smartphones, TVs, cars, sports, and entertainment, with seven major sub-ecosystems, led investors to believe that the synergies of these businesses should not be valued separately but should be priced based on the exponential growth of the overall ecosystem.

No one had ever questioned how much of this 170 billion market cap was at play. LeTV's daily trading volume in 2015 was indeed not low, but of this 170 billion market cap, over 70% of the shares were restricted from trading, and the actual free float was much smaller than the indicated market value. Retail investors and small institutions that could trade drove up the price based on the limited float, which was then automatically multiplied by the total shares to arrive at "170 billion."

A large number is produced → Enters the leaderboard → Provides a sense of certainty → No one has the motivation or ability to question it → A larger number is produced

Thus, Anthropic's "1.2 trillion" is not surprising; it is just the output of the system operating normally.

Anxiety, Anxiety

Let's look at that 1.2 trillion from a different perspective.

What kind of person would buy into a synthetic token with no legal protection at an implied valuation three times higher than the latest institutional round in a market where a liquidity pool is only a little over $1 million?

The answer is someone whose FOMO is so strong that they are willing to pay an anxiety premium for it.

When Anthropic closed its Series G funding in February 2026, the valuation was $380 billion. Two months later, the token's implied valuation on Jupiter was over three times that.

Is this 3x premium due to an information advantage? Are traders on Jupiter more knowledgeable about Anthropic's business than GIC's due diligence team? Clearly not. This premium is not for a cognitive differential but a psychological insurance, hedging against the fear of "what if I miss out."

If you are someone still active in the cryptocurrency space in 2025-2026, what have you witnessed? Anthropic's annual revenue surged from $90 billion at the end of 2025 to $300 billion in May 2026, tripling in three months, a growth rate and amount that most Crypto projects can only dream of. Claude Code's penetration rate among the developer community has exceeded all expectations, while the daily active users of most blockchain applications still hover in the thousands to tens of thousands.

Anxiety is spreading in the crypto community. Are we standing on a ship at low tide, and is AI the true new frontier of this decade?

Jupiter's Pre-IPO token provides an entry point that does not require VC status, a network of connections, or a million-dollar threshold. With just a few thousand dollars, one can hold a stake in Anthropic on-chain.

Buyers from different channels have different anxieties.

Buyers of on-chain tokens are mostly crypto natives. Their anxiety stems from the possibility that AI may be replacing Crypto as the truly disruptive force of this era, while they are stuck in an old narrative.

OTC buyers are mostly traditional primary market participants and family offices. The source of anxiety is that their trading flow is not cutting-edge enough and they are being excluded from the hottest deals. According to Reuters on April 14, Anthropic has received interest from several VCs to invest at an $800 billion valuation, and on April 29, TechCrunch mentioned that the next round might raise $500 billion at a $900 billion valuation. When you don't even have the qualification to receive shares, the panic is particularly strong.

The buyers on the secondary equity platform are mostly U.S. high-net-worth retail investors. The source of anxiety is that every time a tech giant goes public, retail investors are the last ones to get on board, and by the time of the IPO, the upside has already been eaten up by VCs and employees.

But these people have one common trait: they are aware that the premium is extremely high, and exiting is extremely difficult, yet they still buy in because the psychological cost of "missing out" exceeds the economic cost of "being trapped."

Recorded in behavioral economics, the pain of loss is approximately 2.5 times the pleasure of equal gain. But FOMO goes deeper than loss aversion. Loss is a one-time pain, while FOMO is a persistent pain. Humans tend to use the former to end the latter.

So, the most accurate interpretation of that $1.2 trillion figure is: there exists a group of about 3,546 addresses, 298 daily active traders, whose anxiety level is so strong that they are willing to take a risk in a market with just over $1 million in liquidity, offering a premium three times higher than that of a top sovereign wealth fund.

This, in itself, is valuable information. It cannot tell you how much Anthropic is worth, but it precisely measures the temperature of a specific group.

Beyond Price

In a market with thin liquidity, price creates a narrative, the narrative attracts believers, and believers' buying temporarily validates the price. The cycle continues until the number of believers and funds are insufficient to sustain that price.

Each participant has their own logic for decision-making. However, many are likely just ordinary readers who, unaware, internalize the outcome of a simple game involving over three hundred people as their own perception of the world.

Behind the $380 billion valuation from a $300 billion game is a collective pricing with GIC and Coatue putting their reputations as sovereign wealth and top hedge funds at stake. Behind the $1.2 trillion valuation supported by a $1 million daily trading volume is the marginal behavior of over three hundred addresses in an unconstrained market.

From the 1637 Haarlem tavern to the 2026 Solana DEX, humanity has invented numerous market forms, media channels, and financial instruments. However, a problem has persisted for almost four hundred years without a solution, which is how to make the birth conditions of a price—such as how many people, how much money, and under what constraints it was generated—become an integral part of its dissemination.

Perhaps this issue cannot be solved. But at the very least, it should be brought to light.

Welcome to join the official BlockBeats community:

Telegram Subscription Group: https://t.me/theblockbeats

Telegram Discussion Group: https://t.me/BlockBeats_App

Official Twitter Account: https://twitter.com/BlockBeatsAsia