Anthropic Hits One Trillion Overtaking OpenAI

Andthropic's implied valuation on private secondary markets,

specifically on platforms like Forge Global,

has officially touched $1 trillion over taking open AI.

Right, and that valuation,

it comes entirely from secondary market trading.

Okay. So for anyone trying to grasp the mechanics here,

existing shareholders are basically selling their private stakes

to new, usually institutional buyers.

Yeah, so it's not a direct investment into the company.

Exactly. This is fundamentally different

from a primary funding round where the company itself

actually issues new shares to raise capital.

We're going to look at how a company focused purely on business software,

completely outpaced the creator of the most famous consumer

artificial intelligence allocation. Right.

And we will also explore the unique mechanics driving these private

trades, plus the really unconventional ways everyday investors are

trying to get a piece of the action.

So how does a business with a fraction of the consumer user base

justify a higher market value and generate more revenue

than its biggest rival?

Well, anthropic shares are trading

at that implied $1 trillion on Forge Global right now.

Yeah.

Meanwhile, open AI is trading lower,

basically lingering near its last official primary funding valuation.

The demand for anthropic is, I mean, it is so intense

that buyers are literally offering physical real estate

as collateral just to secure shares.

Wow, real estate.

Yeah, you have these massive institutional investors

and highly capitalized family offices,

and they are just scrambling to acquire equity

from absolutely anyone willing to sell.

Right. They just want in.

Exactly.

They are leveraging hard physical assets

to make the math work for the secondary brokers.

It is a completely aggressive posture from the buying side.

And that, you know, that creates a severe supply

and demand in balance in the market.

You have to understand the psychology

and the financial reality of the employees

and early investors who are actually holding these shares.

Yeah, because they're on the inside.

Exactly.

They look at their internal dashboards.

They see the underlying revenue growth trajectory.

And because they see that data,

they fundamentally refuse to sell.

I mean, why would you, right?

Right.

So that creates a market with virtually zero active sellers,

but an endless line of buyers holding giant checks.

When a broker does manage to find like a block of shares,

they are inundated with competing offers almost instantly.

The bidding wars are frantic.

But then you look at OpenAI secondary market

and it's experiencing the exact opposite sentiment.

Of totally.

They have seen a ratio of five sellers to every one buyer.

Five to one.

Yeah, five sellers for every buyer.

That creates a very tepid market for their shares.

Dids are coming in below their last official valuation.

But in Thropic Shares are purchased within hours

of being listed on these exact same platforms.

You really have to wonder about the mechanics

of how these platforms even function

when the demand is that skewed.

Think about a rare art auction.

Imagine the few owners of pieces by a highly coveted artist

simply refused to part with them.

Right, they just lock them in a vault.

Exactly.

They will not sell under any circumstances.

That refusal artificially drive the perceived price up

for the tiny fraction of pieces

that do eventually become available.

Because everyone wants the one piece that's out there.

Yeah, the scarcity itself becomes the primary driver

of the premium.

When there is practically no floating supply of an asset,

the marginal buyer,

the one person willing to pay the absolute highest price,

they set the new price floor for the entire asset class.

I am curious, though,

how much of this is based on real technological superiority

versus just extreme fear of missing out?

That's the big question.

Right, because you have venture capital funds

who feel they absolutely need exposure

to the hottest artificial intelligence company

just to satisfy their own limited partners.

The motivation certainly blends both genuine technological belief

and intense financial pressure.

But the consequence of this behavior is completely undeniable.

This frenzy really alters the perception of leadership

in the artificial intelligence race.

It shifts the focus.

Exactly.

The private market

is heavily rewarding enterprise momentum

over consumer brand recognition.

Institutional capital is sending a clear signal

that deep business to business integration

is fundamentally more valuable

than having a household named chatbot

that everyday people use for trivia.

I think we need to step back for a second.

We are talking about private a-liquid shares here,

not public stock you can just trade from your phone.

True.

But looking at the numbers

and tropics have passed open AI

and annualized revenue run rate,

hitting $30 billion compared to open AI's $24 billion.

That revenue flip happened incredibly fast.

And the underlying economics explain exactly why.

Anthropic monetize is at roughly $211 per user.

Wait, $211?

Yeah, per user.

Compare that to open AI's $25 per user.

And beyond that,

Anthropic boasts over 500 individual customers

who are spending more than $1 million annually.

Oh, wow, so it's heavily enterprise.

Exactly.

They are securing massive long-term enterprise contracts.

Those contracts guarantee consistent capital inflows

rather than just hoping consumers

renew a monthly subscription.

Wait, back up.

How does a software company jump from $9 billion

to $30 billion in annualized revenue in a single quarter?

That math is wild.

That grows driven almost entirely

by high-value enterprise API contracts

and intense developer adoption.

They're not relying on individuals paying $20 a month

to ask a chatbot to write a recipe

or draft a polite email to their boss.

Right, the consumer stuff.

Yeah.

They are integrating their systems directly

into the backend workflows of the Fortune 10.

Imagine you are running a global logistics company.

You do not buy a chatbot.

You purchase guaranteed server throughput.

Because you need reliability.

Exactly.

You pay per token to process massive endless volumes

of proprietary corporate data securely

within your own firewalls.

Every single time your internal software

makes a routing decision, it calls their API,

and they charge you a fraction of a cent.

Multiply that by billions of operations.

Picture the difference between selling

a fleet of commercial jets to international airlines

versus selling millions of bicycles to individual commuters.

Oh, that's a great way to put it.

Right.

The absolute volume of individual users

matters significantly less than the contract size

and the stickiness of the product

within a corporate infrastructure.

A business user utilizing an API for their core supply chain

operations has a significantly higher lifetime value

and a much lower churn rate than a casual consumer

testing out conversational software.

This proves consumer virality is absolutely not

the only path to massive software revenue.

It fully validates the business-to-business enterprise model

as the most lucrative route in this entire sector.

Yeah, definitely.

Getting to 100 million users the fastest

was a phenomenal headline for consumer apps.

But deeply embedding your models into the software

that actually runs global finance, healthcare systems,

and logistics networks.

Well, that is what generates $30 billion

in highly reliable recurring revenue.

And a specific developer tool called Cloud Code

is generating $2.5 billion in annualized revenue

all by itself.

Just that one tool.

Just that one.

That agentic coding tool is now

authoring a significant percentage

of all public code commits on GitHub

with internal projection showing it could handle

a fifth of all commits globally.

A fifth?

That is unbelievable.

We are looking at a single developer product

that's scaled from zero to multi-billion dollar revenue

in a matter of months.

Entire engineering teams at major corporations

are utilizing this tool to automate their routine coding

tasks, completely debug complex legacy systems,

and heavily accelerate their entire software

deployment pipeline.

See, I look at that and see a massive concentration risk.

Really?

How so?

Well, relying so heavily on a single product

category like coding assistance makes

the overall company highly vulnerable.

If well-funded competitors release highly capable open-source

alternatives, that entire $2.5 billion revenue stream

could evaporate overnight.

You think they would switch that fast?

Absolutely.

An enterprise client will happily switch

to a free or heavily discounted alternative

if the performance gap between the tools narrows.

Code generation is actively becoming commoditized.

I disagree completely with that assessment.

Developer tools have the highest switching costs

in the entire software industry.

But if it's cheaper?

Doesn't matter.

Once you have a development team,

integrate an agentic coder into their daily deployment cycle.

It becomes entirely entrenched.

You establish your core security protocols

around this specific tool.

You spend months trading your software engineers

on its highly specific nuances and prompting structures.

I guess the training does take time.

It takes a lot of time.

It creates the deepest possible entrenchment

in enterprise workflows.

Removing that tool disrupts your entire engineering pipeline

and completely kills internal productivity.

No chief technology officer will risk halting

their product roadmap just to save a few million dollars

on software licenses.

That dynamic absolutely shifts the focus

of modern software development then.

Developer focus tools are actively establishing themselves

as the most profitable layer of the new software stack.

Exactly.

This opens up entirely new categories of enterprise spending.

You are seeing chief technology officers

aggressively reallocating massive budgets

from other departments just to equip their engineers

with these specific capabilities.

The return on investment in pure labor efficiency

is undeniable.

If a tool makes your engineering team 20% faster,

you pay whatever the vendor asks.

And here's another wild stat.

Anthropic projects spending four times less

on model training than OpenAI.

Four times less.

Yeah.

OpenAI is projected to spend over $120 billion

on compute over the coming years.

While Anthropic projects their spend at around $30 billion.

That is a massive gap.

It is.

And because of this discipline spending

and highly focused engineering strategy

and theropic projects reaching actual profitability

much sooner, they are optimizing their model architectures

to achieve top tier reasoning capabilities

without requiring the same brute force computational power

that their competitors rely on.

Hold on, does spending less naturally

mean their models will eventually fall behind in performance?

Not necessarily.

I mean, the entire industry currently

assumes pure size and scale are the only mathematical ways

to achieve advanced intelligence.

Cutting costs on training runs

seems like a guaranteed recipe for losing

the capability race in the long run.

Anthropic is aggressively testing the assumption

that the company spending the absolute most on training

automatically wins the market.

So they think there's a smarter way.

Right.

They are betting heavily that algorithmic efficiency,

high quality synthetic data generation,

and highly specialized model routing matter,

significantly more than just raw compute volume.

They believe they can match or exceed

competitor performance through smarter, highly targeted

engineering rather than just buying more graphics processing units

and plugging them into the wall.

I am still highly skeptical of the math

on the other side of that equation.

You have to wonder if these companies

are just throwing billions of dollars into a bottomless pit.

It definitely looks like that sometimes.

Building an artificial intelligence model right now

feels exactly like building a massive, highly expensive

highway system before knowing exactly how many cars

will actually use it.

You commit tens of billions of dollars

to massive data centers, cooling systems,

and specialized silicon chips.

Just hoping the resulting model produces enough commercial value

to justify the massive capital expenditure.

That changes how institutional investors

evaluate these companies entirely, though.

Financial success is increasingly measured

by who can generate the most revenue

per single dollar of training spend,

rather than who simply builds the biggest underlying model

with the most parameters.

It's all about efficiency now.

Think about the pure math of it.

If you spend $100 billion just to train a single model,

your revenue requirements to reach free cash flow

are astronomically high.

Capital efficiency is rapidly becoming the defining metric

for long-term survival in this specific sector.

Let's talk about where that money is coming from.

Major technology giants provide enormous funding

to artificial intelligence startups, who then turn around

and use that exact same money to buy cloud computing services

from those exact same tech giants.

It's quite the loop.

Amazon and Google have made multi-billion dollar investments

directly into Ampropic.

And that creates a highly constrained circular financing loop.

Amazon invests billions of dollars

into the startup's bank account.

But Ampthropic explicitly commits

to using Amazon Web Services for its massive compute

needs to train and serve its models.

So the money just goes right back.

Yeah, the capital effectively never

leaves the corporate ecosystem.

It is simply reclassified from an investment asset

on the balance sheet to cloud computing revenue

on the income statement.

Big tech gives them money, and they hand that money

right back to Big Tech to rent servers.

It's that simple.

It operates exactly like a commercial landlord

giving a tenant a large cash loan specifically

so the tenant can pay the monthly rent directly

back to the landlord.

That sounds incredibly fragile.

You have to strongly question the long term health

of this financial dynamic.

If the underlying enterprise return on investment

does not materialize for the actual end user buying

the software, this entire closed loop

could collapse entirely.

The major cloud providers are heavily subsidizing

the artificial intelligence boom specifically

to drive their own infrastructure growth metrics

for Wall Street.

This arrangement completely limits

the true independence of these artificial intelligence

startups.

But I guess it absolutely ensures

their short term survival.

They couldn't survive without it.

It guarantees immediate access to crucial computing

infrastructure.

Without these highly strategic partnerships,

securing the necessary gigawatts of electrical power

and hundreds of thousands of specialized chips

would be mathematically impossible for an independent startup.

They are trading complete autonomy for access

to the only machines capable of running their software.

Now here's a really strange angle.

Zoom video communications owns a stake in Anthropic

that could be worth up to $10 billion.

Wait, Zoom, the video call company.

Zoom, they made an early $51 million investment

directly into Anthropic.

Even factoring in the heavy share delusion

from subsequent massive funding rounds,

a 1% stake in a trillion dollar company

mathematically equals $10 billion.

That is wild.

Buying Zoom stock right now operates

as a brilliant, highly asymmetric bet

on Anthropic's future success.

Zoom is currently sitting on nearly $8 billion

in cash and short term investments with absolutely zero debt.

So it's like a back door.

Exactly.

If you buy Zoom equity,

you are effectively getting Zoom's core video conferencing

business completely for free,

simply given their cash reserves

and the enormous paper value

of that artificial intelligence equity.

I have to push back on the idea

that this is a safe investment vehicle though.

Zoom's core business faces extreme competition

and steadily declining growth.

Sure, but the cash buffer.

But think about it,

the permanent shift in hybrid work models,

combined with intense pressure

from bundled enterprise software solutions

like Microsoft Teams,

puts heavy downward pressure

on their enterprise retention metrics.

Who's a fair point?

That makes it a highly risky vehicle

just to hold an artificial intelligence lottery ticket.

If Zoom's enterprise market share arose further,

the equity value of the core business

will drag down the entire stock price

completely regardless of what the underlying

and through up stake is actually worth.

I see what you mean.

But this opens up a highly creative strategy

for everyday retail investors

to gain exposure to private market darlings.

They do not need the legal accreditation

or the massive capital required

to buy private shares directly.

They just buy Zoom instead.

Right.

Retail investors are actively hunting

for publicly traded proxies

to capture the financial upside

of companies that remain strictly locked

behind private venture capital doors.

They are looking for any backdoor entry

into the wealth creation event.

And some are going even further.

Tokenized anthropic shares are actively trading

on cryptocurrency platforms,

implying an $850 billion valuation.

The crypto angle is fascinating.

These highly experimental decentralized pre-stocks,

trade on networks like Salana,

theoretically allowing everyday retail participants

to speculate directly on the private market valuation.

But there is a critical structural flaw

in that specific trading system.

A retail trader made over $1 million

in pure paper profit trading these tokens.

But they cannot actually cash out

due to a severe lack of market liquidity.

Wait, so they have a million dollars on screen,

but can't touch it.

Exactly.

When they actively attempted to simulate selling their position

across various decentralized exchanges,

the financial slippage was massive.

There were simply no actual buyers sitting

on the other side of the trade at that specific price point.

Wait, let me understand this fully.

People are purchasing digital tokens

that represent private shares,

but they do not actually own the underlying shares at all.

Right, they are purely bare digital assets.

They carry absolutely no legal voting rights.

They provide zero direct ownership in the company itself.

So what a day.

They are highly complex synthetic instruments,

explicitly designed just to track

the price of the underlying private asset.

They are managed by highly obscure special purpose

vehicles that hold the actual legal paperwork offline.

You are basically holding a winning lottery ticket

in a jurisdiction where you cannot legally claim

the financial price.

The numbers look phenomenal on a digital screen

and the implied valuation creates highly clickable headlines.

But the actual real world financial value is effectively zero

if you cannot mathematically convert those digital tokens

back into usable fiat currency.

Yeah, the liquidity trap is absolute.

This completely limits the actual value

of these synthetic assets.

It heavily highlights the extreme highly risky lengths

everyday retail investors will go to for any exposure

to the artificial intelligence financial boom.

People just want in so badly.

The pure desperation to participate

in this specific wealth creation event

is driving retail capital into highly experimental

and intensely illiquid financial structures.

People are buying complex derivatives

just to feel like they are participating.

And all this pressure is building up to something.

Both open AI and anthropic

are aggressively preparing for initial public offerings,

heavily driven by competitive market dynamics

and mounting employee stock options.

There is an intense internal pressure cooker building

inside these exact companies.

Thousands of highly paid engineers and researchers

are holding illiquid stock options

that strictly vest over a four year period

the standard one year cliff.

And they want their money.

Now of course they do.

These individuals have generated billions

in theoretical paper value.

They naturally want to realize

the actual financial gains of their incredibly hard work

to buy homes and permanently secure

their personal financial futures.

Investment banks predict these eventual public market offerings

will heavily exceed $60 billion in raised capital.

The scramble among major global financial institutions

to secure the rights to underwrite these listings

is intensely competitive.

I fully believe the public markets

will eagerly absorb these specific offerings

simply due to the pure revenue growth velocity

we discussed earlier.

You think so?

Yeah.

When massive mutual funds, global pension funds,

and everyday retail investors finally secure

a direct mechanism to buy the company's

actively defining the next computing platform,

the capital inflow will be absolutely staggering.

The public market is entirely starved

for pure play, artificial intelligence,

infrastructure stocks.

I completely disagree with that outlook.

Because public markets strictly demand clear,

mathematically sound paths to free cash flow.

Wall Street analysts will absolutely not

stomach revenue multiples of 27 times or higher

for companies that are actively burning billions of dollars

every single month on raw server costs.

But the growth is unprecedented.

The transition from private venture markets

to public markets involves a highly brutal shift

in financial valuation metrics.

Private venture capitalists value exponential

top-line user growth above all else.

Public market analysts strictly demand highly sustainable

operating margins and incredibly clear profitability timelines.

This sets up the absolute ultimate test

for the entire artificial intelligence industry then.

The harsh transition from private venture capital logic

to strict public market scrutiny

will completely force these specific companies

to legally prove their underlying business models

are highly sustainable.

They will have to mathematically demonstrate

that they can sell digital intelligence

to enterprise clients at a strictly higher margin

than it natively costs to generate it on their hardware.

So tying this all together, the competition

between artificial intelligence leaders

has entirely shifted from consumer popularity

contests to a ruthless focus on enterprise revenue

efficiency, strategic cloud partnerships,

and intense private market maneuvering.

If the physical hardware and daily compute costs

continue to rise exponentially,

will these trillion dollar valuations

hold up when the public markets finally

demand actual sustainable profit margins?

If you're not subscribed yet, take a second,

and hit follow on whatever app you're using.

It helps us keep making this.

We appreciate you being here.

Also check out our YouTube channel for more business

and tech updates.

There's a link in the description.