Author: Fu Sheng

Previously, we discussed the strengths and weaknesses of Anthropic and shared three records set by Anthropic.

This Monday, there was a major development in Silicon Valley's AI circle, as Anthropic took the lead over OpenAI and secretly submitted an IPO application. Morgan Stanley and Goldman Sachs are the joint lead underwriters, with the quickest listing expected in October this year.

After completing a $65 billion Series H round, its valuation is at $965 billion — just a step away from reaching one trillion.

When it finally goes public, reaching a trillion is inevitable, and there's even a possibility of hitting 1.5 to 2 trillion. If it really hits 2 trillion, it will surpass SpaceX to become the highest-valued company in the Pre-IPO stage worldwide.

Many friends have left messages saying "Fu Sheng, aren't you exaggerating?" A startup founder even directly asked me: When will this AI bubble burst?

My answer has always been the same: If you understand the situation, you won't see it as a bubble.

Is it really a bubble?

The founder's logic is straightforward: Which company can be worth $1 trillion just four or five years after its establishment? Didn't the internet bubble burst in 2000 for similar reasons?

I said not to rush to conclusions. The term bubble has been thrown around so much that no one has seriously considered the question: What is similar and what is different between the bubble of 2000 and today? If you draw out these two lines and compare them, the answer will reveal itself.

Similar history does not mean history is repeating

If you say that just because the Nasdaq crashed in 2000, AI in 2026 must have a bubble, then that line of thinking needs an upgrade.

What is similar? It’s the emotions. A new technology rises, capital floods in, valuations soar, outsiders are confused — this psychological script is indeed reminiscent of those years.

But it’s the differences that are key to judging whether there’s a bubble. If you only look at the first half and jump to conclusions, that’s like trying to find a sword in a boat.

Telling stories and making calculations are two different businesses

What did internet companies rely on for valuations in 2000? A domain name, a PowerPoint presentation, a "market dream ratio." There were no revenues, no profits, and even no paying customers, purely supported by stories and imagination for stock price. At that time, analysts discussed "how big this company might become," and that "future" had no revenue data to support it.

Today, Anthropic's situation is clear.

We’ve previously discussed that Anthropic is currently the fastest-growing company in revenue, with the highest output per person and the highest valuation in human history for private companies.

Let’s first look at revenue growth. Its ARR was $1 billion at the beginning of 2025. By the end of 2025, it reached $9 billion. By May this year, it had soared to $47 billion. Internal documents show the target for year-end is $100 billion — going from $1 billion to $100 billion in under two years. No other company in business history has achieved this trajectory.

Moreover, this isn’t just a result of burning cash; Anthropic is expected to achieve a single-quarter revenue of $10.9 billion in the second quarter of this year, marking its first operational profit in history, roughly $560 million, already starting to make money before going public.

Next, let’s look at output per person. Anthropic currently has about 3,000 employees, and with an annualized revenue of $47 billion, the output per person exceeds $10 million.

A single programmer paired with Claude Code can accomplish work equivalent to that of an entire team. Moreover, this Claude Code product, which has been on the market for less than a year, has achieved annualized revenue of $2.5 billion from just one product, capturing 54% of the AI programming market.

Finally, let's review the valuation logic. Anthropic's main business is selling API subscriptions to enterprises. If you look at it as a traditional SaaS company, there's a classic formula for valuing SaaS companies: the Price-to-Sales ratio equals market capitalization divided by annualized revenue, multiplied by a multiplier. Subscription revenue from enterprise customers is very stable; as long as the renewal rate is over 95%, the capital market generally gives it a 10x multiplier. If it really achieves $100 billion by the end of the year, 10x would mean a market cap of one trillion.

You might argue that this valuation model is itself not very reasonable, but you cannot say that it lacks a model. During the internet bubble, pricing was based purely on imagination. Today, Anthropic’s financials are laid out right in front of you. Its client list is also solid: eight of the top ten from Fortune are using it, and over a thousand major enterprises spend over one million dollars annually on Claude. Netflix, Spotify, KPMG, Salesforce, all are on the list.

Back during the internet bubble, analysts asked, "How big can this company become in the future?" Today, analysts are asking Anthropic, "How much did you earn this quarter, and how much more can you earn next quarter?" This is the fundamental difference.

From Carbon-Based Economy to Silicon-Based Economy

After discussing these two reasons, I still feel it's not enough. Behind the listing of Anthropic lies a larger, deeper trend: human economics is transitioning from a carbon-based to a silicon-based model.

Recently, in an internal meeting at the company, I said we should rename the HR department to IR, changing from Human Resource to Intelligence Resource. A company’s intelligence level and competitiveness now depend not only on how many people and smart minds it has but also on how much computational power, how many models, and how much AI capability it has to scale repetitive tasks.

This concept isn’t something I invented; companies in Silicon Valley have already validated this logic. Bryan Catanzaro, Vice President of Deep Learning at Nvidia, publicly stated two months ago that his research team has entered the stage where "computational power investment surpasses human input," with the cost of computational resources far exceeding employee salaries.

Recently, Sam Altman also mentioned an interesting phenomenon: many enterprises used up their entire AI budget in Q1 this year. Companies have realized that every dollar spent on model capabilities and computational power directly enhances product competitiveness—development efficiency, customer response speed, data analysis depth, these aspects which used to be achieved by simply piling up manpower are now being scaled up via AI.

The human economy is shifting from being solely carbon-driven to being driven by both carbon and silicon, which is the real change happening behind Anthropic's listing. It is not just a company ringing a bell; it represents a price anchor for a new economic paradigm.

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。