Anthropic Valuation Nears $900 Billion
A deep-water bomb in the AI world.
On April 29, it was reported that Anthropic is negotiating a new round of funding, with a valuation that could exceed $900 billion.
If the deal closes, this company—founded less than four years ago—will surpass OpenAI in one leap, becoming the most valuable AI unicorn on Earth.
$900 billion.
What does that number mean?
In China’s A-share market, it’s higher than Kweichow Moutai’s market cap. In Silicon Valley, it kicks over the valuation throne that OpenAI spent a decade building.
More intriguing is the timeline. Just a few months ago, Anthropic was valued at $60 billion.
Google and Amazon invested at a $350 billion valuation, committing a combined total of up to $65 billion. From $60 billion to $900 billion, less than a year—a 15x increase.
Capital doesn’t lie. When the smartest money in the world rushes in the same direction, something has definitely changed.
So what changed?
The answer is striking—
Users Are Voting With Their Feet
According to market intelligence firm Sensor Tower, ChatGPT’s uninstall rate in April increased 132% year-over-year. The month before, it was even higher—up 413%.
At one point, ChatGPT’s uninstall surge hit 563%. During the same period, Claude’s downloads jumped 199% in a single week.
In multiple countries, Claude reached the top of the iPhone free app charts.
In the U.S., Claude’s daily downloads surpassed ChatGPT for the first time. On February 28, it climbed to #1 on the App Store free ranking and held that position until March 2. Within a week, it jumped more than 20 spots. Claude also topped the iPhone free app rankings in Belgium, Canada, Germany, Luxembourg, Norway, and Switzerland.
Users are voting with their feet. And they’re pressing the gas pedal while doing it.
OpenAI: Cracks Begin From Within
On the surface, OpenAI still looks massive.
The GPT series has hundreds of millions of users. Codex has just sparked a new wave of excitement. The Stargate project claims a $500 billion investment to build AI infrastructure.
But empires often crack from the inside first.
If user loss is a surface wound, then the downsizing of “Stargate” is a broken bone.
According to reports, the actual progress of Stargate is far less impressive than its slides suggested.
$500 billion. Ten nuclear plants. The only path to humanity’s future. Now it looks more like a downsized rental contract.
The UK project—halted.
The Norway project—cut.
The Texas flagship base—abandoned.
Altman calls it “flexibility.” Partners call it “pulling the rug out.” SoftBank is frustrated. Oracle is recalculating. Microsoft is quietly picking up the pieces.
The subtext is clear: once OpenAI starts stepping back from infrastructure responsibility, it loses control over the physical world.
Funding timelines, data center locations, partner coordination—every step is dragging.
“Progress is much slower than expected.”
Talent Drain in Silence
At the same time, OpenAI is experiencing a quiet loss of talent.
Dario Amodei—Anthropic’s founder and CEO—was formerly OpenAI’s VP of Research.
He didn’t just leave himself. He took a group of OpenAI’s most core safety researchers with him.
Speaking about his departure, he said plainly: “Instead of staying to argue over someone else’s vision, it’s better to take people you trust and build your own vision.”
This kind of exit hasn’t stopped.
Over the past two years, key members of OpenAI’s alignment and safety teams have continued moving to Anthropic.
What’s the most valuable asset of a company?
Not user numbers. Not valuation. It’s the people who can define the next generation of models.
When those people choose to leave, the direction itself becomes the answer.
Anthropic: From “Safety Lab” to “Most Valuable Unicorn”
Anthropic’s rise doesn’t follow the usual Silicon Valley script.
It didn’t start by burning money to grab users.
It started with a paper on AI safety. Its core idea was “Constitutional AI”—using a constitution to constrain model behavior.
For a long time, mainstream opinion in Silicon Valley was: these people are too idealistic, they won’t scale.
Then Claude 3.5 Sonnet was released.
Its coding ability crushed GPT-4o. Its long-context understanding pulled far ahead. Hallucination rates dropped significantly.
Overnight, the developer community shifted.
Reddit, Hacker News, X—everywhere you saw the same sentence:
“I canceled ChatGPT Plus and switched to Claude.”
Not one person. Thousands of people.
Enterprise Shift Toward Claude
More importantly, the enterprise side.
On AWS, Claude’s API usage has more than doubled in the past six months.
More and more companies are migrating core business from GPT to Claude—not because it’s cheaper, but because it works better.
Google saw this and invested. Amazon saw it and invested too. Together they committed $65 billion—not betting on Anthropic’s present, but on its ability to define the next generation of AI.
Is OpenAI Really Losing?
Not so fast.
GPT-5 continues to iterate at an astonishing speed. Version 5.5 just landed, and traces of 5.6 are already appearing in backend logs.
Codex, as an agent tool, is taking off across the board. The developer ecosystem moat isn’t something that can be broken overnight.
Another card OpenAI holds is scale.
Hundreds of millions of monthly active users. Deep integration with Microsoft. Enterprise customers worldwide.
These advantages don’t evaporate because of one funding headline.
History shows that surpassing valuation doesn’t decide the winner.
When Facebook went public in 2012, many thought Google’s social ambitions were finished. Ten years later, Google’s search empire was never truly threatened.
The brutality of the AI race lies in this:
Today’s throne might not count after the next model release.
A Shift in Power Structure
Pull back, and this isn’t just a story of who wins or loses.
It’s a story of the AI industry’s power structure being rewritten.
Two years ago, OpenAI was the only superstar.
One year ago, Google’s Gemini started catching up.
Today, Anthropic has reached the top in valuation.
Meanwhile, xAI is burning cash, Meta is stirring the pot, and open-source AI is closing the gap.
The winner is no longer singular.
AI is moving from dominance by one to fragmentation among many.
The Only Variable That Matters
In this chaotic battle, only one variable truly decides the outcome:
Who can build the next-generation model first.
Not who has better slides. Not who raises more money.
AI isn’t magic. It’s heavy industry.
It needs astronomical electricity. Absolute credibility. Ruthless financial discipline.
Anthropic’s $900 billion valuation is, at its core, a market bet:
The next “iPhone moment” might not be in OpenAI’s hands.
