Hacker News new | past | comments | ask | show | jobs | submit
I am totally blown away that a company that was formed in 2021 will have a 1.5 trillion market cap on the public market... to put into context, a valuation not achieved by Apple until 2020
It doesn't say much about the company itself, but about the state of the economy, where all the money has been concentrated at the top, in the hands of people who have nothing else to spend it on than on more financial assets. So there's enormous amounts of money available for more stocks.

It's why stock markets and corporate profits are at an all time high, while most people struggle to make ends meet and governments are woefully underfunded.

That $1.5 trillion IPO is the result of a completely lopsided economy.

The K shaped economy trope is directionally correct, but you have the cause backwards.

It's not that governments are underfunded, the K economy is a direct byproduct of government stimulus across all developed economies for the last 15 years.

When you inject the market with QE, the money printer and most importantly; 0% or negative interest rates for 15 years...it massively inflates the value of assets. Rich people own assets.

This is why things in the US with the prospect of generating future cash flows are at historically high valuations relative to GDP: public companies, private companies, real estate, etc.

If you think government is 'underfunded' and the solution is giving it more money to dump into the economy, ironically, you're going to inflate asset values further and make the K economy worse.

The stimulatory tools that are easiest for government to use are monetary policy, not fiscal policy.

But let's say you do manage to pass populist fiscal policy, like, giving first time home buyers $20k to help with a downpayment on a house.

Surprise! Turns out you've simply raised the value of all starter homes by $20K by increasing everyones capacity to pay for a scarce asset, inflating the value of the asset and handing the money to the person selling it.

The solution is to increase the supply of assets (build more real estate and competing businesses), not stimulate demand for existing ones. But that requires a much more rational, long term policy view than dumping blankets of cash on angry populist voters.

You are describing, almost fairly well, issues of accumulated wealth at the top -- particularly the effects of covid handouts[1], where all the money ended up with landlords that were not allowed out to spend it all, significantly increasing inequality and asset prices.

And somehow you try to turn it into an anti-tax argument, when taxing the wealth at the top is the only solution, the way to control the money supply, time machines aside. Give out too much you have to take some back if you don't want to devalue.

1. which is why the stock market and housing market went crazy at that point, not a steady 15 years, $5tn was created

> Give out too much you have to take some back

This would be true only if the government simply increased taxes on assets (say, doubled capital gains taxes) and then took that money and paid down government debts with it.

However, we all know this won't be the case. The money taken from asset holders will simply be dumped right back into the economy via populist handouts, and that money will ultimately just re-inflate the things that are scarce and valuable (assets).

You're disincentivizing the creation of assets at the individual level (by taking them from people) and stimulating demand for them at the societal level by giving everyone money to chase assets with. A recipe for more runaway prices.

Again, there is only one path: increase the supply of assets (build more real estate and more competing businesses). China has similar/lower tax rates than the US in most areas while their middle class grew massively. This is how. It's not magic.

loading story #50561567
It doesn't matter how all of the money ends up at the top. Yes, government handouts to the rich are a big part of the problem, but so are a lack of taxation on the rich and on corporations, a lack of competition, the fact that both corporate profits and inflation are high, and the lack of stimulus for the poor.

You seem to think that the government can only use its money for more stimulus for the rich, but that assumption is itself part of the problem.

If the government gives more money to the rich, whether through tax cuts, subsidies, handouts, or whatever, the "trickle down" lie has always been that that money will eventually help everybody, but that's only the case if that money ends up in the economy, paying wages, etc. But instead, that money is being used to buy up more assets. That doesn't help the economy in any way, it just skews it.

If instead, that money ends up in the pockets of people who need it, they will immediately spend it, buy food, buy other things they need, pay off debt, and all of that money creates demand, pays wages, and grows the economy in a way that helps everybody, and doesn't just inflate the numbers.

Alternatively you could massively tax the wealth of the very rich and revert the concentration. Then you are not competing with them for the scarce assets.
You could take every dollar of wealth Elon has and give every American $2,500. Is that really going to help?

More importantly, taxing the rich based on unrealized wealth means forcing them to sell their assets. Markets would be flooded with new supply and no buyers, tanking prices and putting everything in a much worse position.

loading story #50561739
loading story #50559126
Again, giving government more money to dump into the economy will do literally nothing to solve the problem.

You will simply be taking money from the rich, and then giving it right back to them by inflating asset values.

Because the assets are the things that are scarce and valuable, and you're now disincentivizing the creation of them while stimulating demand for them more by dumping cash on everyone.

There is only one path: creating more valuable things (more supply of real estate and businesses offering competing products lowers their price). China, ironically a 'communist' country, figured this out yet we're still struggling to.

The truth is we don't want to face this reality. We'd rather sit at computers, place fantasy expectations on the politician du jour to save us, and repeat smart sounding memes we don't understand for magical internet upvotes ('late stage capitalism' 'K shaped economy') instead of picking up a hammer and building anything.

loading story #50560119
loading story #50562037
loading story #50559178
Further its because income generated by assets are tax-efficient in the western world (well at least North America) in a way that non-asset generated income (i.e labour income) is not. Naturally the inclination is to not just accumulate assets for their own worth but for more tax efficient income as well.

If asset-based income was taxed similarly or included in the tax base, you wouldn't have stimulus become inflationary in the way it is today.

> "The solution is to increase the supply of assets..."

Let's be concrete (and unpopular): If one of the few assets left to buy for billionaires is land, which creates competition among them for land not to live on but to HOLD and rent out, the solution is not to "increase the supply of land", it's to decrease the amount of billionaires...

Land is not actually that scarce. Land in cities is scarce.

We can build more cities. How do you think the current cities we have came to be?

Yes, it requires a longer term view than a 4 year election cycle. China has created hundreds of cities from nothing in the past 30 years. Many failed, many succeeded. This is the risk taking required to create new things.

We could do it too. But we choose to stagnate and inflate the real estate we already own instead.

loading story #50558806
> It doesn't say much about the company itself, but about the state of the economy, where all the money has been concentrated at the top, in the hands of people who have nothing else to spend it on than on more financial assets.

Huh, what?

> So there's enormous amounts of money available for more stocks.

Money (nor stocks) don't have to change hands for stocks to become more valuable. All that's required is that the likes of Citadel and Jane Street quote higher prices, no trading necessary. (Of course, they can and will only quote higher prices, if that's justified by supply and demand.)

> It's why stock markets and corporate profits are at an all time high, while most people struggle to make ends meet and governments are woefully underfunded.

Government expenditures have grown like crazy in much of the world, haven't they?

That's not how supply and demand works. There has to be trade in order to establish a price.

And governments in much of the world have actually been cutting expenditures, leading to serious shortages in many places. Increased poverty, reduced quality of education, crumbling infrastructure. Governments were better at this from the post-war period to roughly the 1990s, until neoliberalism and austerity became the law of the land.

> That's not how supply and demand works. There has to be trade in order to establish a price.

Why? Anything you can potentially trade against, especially with a tight spread, counts in my books as a price.

> And governments in much of the world have actually been cutting expenditures, leading to serious shortages in many places. Increased poverty, reduced quality of education, crumbling infrastructure. Governments were better at this from the post-war period to roughly the 1990s, until neoliberalism and austerity became the law of the land.

https://en.wikipedia.org/wiki/Austerity says

> Austerity is a set of political-economic policies that aim to reduce government budget deficits through spending cuts, tax increases, or a combination of both.

Government deficits have reached record highs in eg the US and France. Where is that austerity?

Neoliberalism is, alas, dead. The glory days of the 1990s and early 2000s died around 2008.

Would upvote twice if I could.

Another channel for the wealthy to place the money they don't know where to spend anymore. And the audience applauds

Also, it's a valuation for selling to asset holders, no such profit has been achieved yet with the actual portfolio.

It's like an IPO for a beloved artist who claims he will sell a painting to every household in the world.

> "It's like an IPO for a beloved artist who claims he will sell a painting to every household in the world."

So not that far away from the usual "let's all buy from this artist and push his works to a billion dollar" scheme the wealthy usually follow.

But now with tech...

Well at least profit from the appreciation of artworks can be written off a little easier. Stocks are too liquid and regulated to serve the same purpose.
loading story #50558286
Who knew that having wealth accumulate instead of flow would be bad for an economy that relies on the allocation of capital? Every economist ever.

Billionaires are the economic equivalent of a tumor. I don't know why we're so obsessed with perpetuating them.

This is an interesting position.

high volume, 'flow', of wealth is usually what economists call a hot economy, which usually leads to inflation which then usually leads to problems that a central bank has to mitigate or consumers have to bear the weight of.

I know this is getting into the weeds of economics ...the OP was posting about a pending IPO...just setting something out for consideration....

There is something more though I think in terms of lack of cultural capital.

There isn't enough cultural activities we value in society so we have largely turned our culture into this kind of net worth sporting contest. Billionaire is just round number bias that reflects a goal post in the sport at the highest level.

Late neoliberalism I believe is the right frame but this is still a rather obscure idea.

It's annoying that people all readily acknowledge that billionaires do not have piles of cash, but then they incessantly talk as if they do.

My friend, the billionaires wealth is the economy. It is the thing that is flowing the money.

Jeff Bezos wealth isn't cash redeemable pieces of paper with "Amazon Share" stamped on them.

His wealth is warehouses filled with goods, software running logistics networks, CPU's serving web services, trucks delivering packages, and a workforce of 1.5 million people all aligned on serving paying customers for their needs or wants.

So while you can detest it and hate it, at least recognize that that is what you call "accumulated wealth", not a pile of money where we are one legislative act away from busting open like a pinata.

If you snapped your fingers to distribute his wealth, you'd feel pretty dumb when you end up with 13 aluminum dolly carts as your share.

loading story #50562402
Because billionaires and those right below them spend a ton of capital on perpetuating the idea that the average person can be like them. Hell they even have their own podcasts now lol
> "...that the average person can be like them"

...or become a beneficiary of the benevolent billionaire.

My favorite quote comes to mind: "The wood will not vote against the axe, because it has unrealistic expectations to become an axe-handle one day"

It's an absolute state when hackernews from all places has degenerated into braindead leftist "economics" echo-chamber.
I love how the exact same people I describe above (normal people who've be brainwashed to think they can achieve the same level of success that only ~3k billionaires have had), have also been programmed to automatically dismiss anybody that writes a sentence that uses the word billionaire negatively, as "braindead" leftist economics.

They have brainwashed you to the extent that you spend your time reflexively defending them in the comments of online spaces. You become an activated foot soldier, coming to the defense for the very people who seek to enslave you the second anyone says something bad about them.

You proved my point.

Nope, I just understand basic economics and have logical deduction skills instead of buying sentimental bullcrap from demagogues. Please read a book, you will be doing all of us a favour.
What book do you think would change my mind?
For your level even basic economics books from any 101 course in a university (google away) would help you a lot.
I took many economics classes, its partially why I have informed opinions on the matter. You're so confident reading a book would change my mind, surely you have a reccomendation.. Or are you just politicing? Maybe lay off the X algo.
I don't have an X, never did. Was it a Learing Center you got your economics classes from?
Perhaps you should be the one to read that book. None of the things discussed here are in any way controversial, except with the billionaires and their politicians. Many economists are saying the same things we are saying. That's where we got it from.
The K shaped economy is developing serifs.
It's K shaped economy with the upper arm extended with additional /
We're going to need an additional Unicode character for that.
Well, yes. But also the company is projecting to sell pretty much every white collar worker a $200/mo Claude subscription, which is an absurd amount of revenue compared to said Apple, which promises to sell a new phone to 20% of people every few years.
But will it be Claude? Probably it goes through Microsoft through copilot. In which case it becomes a race to the bottom. I never need Fable even today: Opus suffices on 99%. So why do I need anthropic? Opus will be replicated in half a year by every lab.

In that sense I am super bullish on Microsoft as they have the vendor lock.

For anyone who has used the various tools, the harness still matters and copilot is pretty bad. While Anthropics models lose on cost compared to OpenAI, there is still something that works really well in Opus paired with the Claude desktop app when it comes to tool use and solving tasks.
I see what youre saying, but there is no way a TUI or GUI is anthropics 1.5T moat
Why are you pivoting the argument? You said why use Anthropic because copilot can use the models. For anyone who has used copilot and Claude, they are worlds apart and MSFT has been working on copilot for more than a year and they are light years behind and are not answering simple business use cases. I am not here defending the value of their moat but simply the harness still matters.
"compared to said Apple, which promises to sell a new phone to 20% of people every few years."

- around 50% in the US market

- they offer way more products, than phones, are the leader with laptops for instance

- take a big cut of every store purchase

The US is just ~350 million out of the ~8000 million people on the planet.
It's hard to put a percentage on it, but I'd wager for 90% of the earth's population, paying $200 a month for a personal LLM subscription isn't viable or simply not worth it.
When something like 60% of the world’s population lives on less than $200 a month, it’s a preposterous suggestion.
What % of those 8B people can afford an iPhone? How many of those people are located in the US?
Uh, yeah of course a person is super rich for having $1000 or $100 invested in the stock market.

No, of course a person is not super rich for owning their own home which is worth hundreds of thousands.

God damned young yuppies with their stock portfolios worth $20 000 or more! In the meanwhile grandpa is suffering in his house which is only worth $1,3 million.

If only we could take the money out of investment into businesses and put them into investing in real estate! That's where all the jobs are made who pays people's salaries. Can somebody call the EU to have them regulate?

I am totally blown away that a company who does not turn a profit, whose revenue is slowing down, and has hard financial commitments 25x times, the money they just raised, over the next four years, was able to get 81 billion from investors.

As predictions are only worth anything, if they are made ahead of time...I predict their share price will lose 25% of its value,over the next 3 months. Drop to 40 to 45% of IPO price, after their first public quarter release.

  and has hard financial commitments 25x times, the money they just raised, over the next four years,
They have about $1 trillion commitment in compute over the next 5 - 15 years. I'm not sure where you got the 25x$81b from.
But most is concentrated in the first 3-5 years. Is pretty bad for their cash flow.
The first 5 years estimate is around ~$355b–419b. We don't know for sure because most of their deals actually span 5+ years, up to 10-15 years.

They're at $86 annualized revenue right now. If they grow at 0% for next 5 years, they will generate $430b, which is enough to cover all compute costs.

$1t notional / contractual but the negotiable terms are unique
> As predictions are only worth anything, if they are made ahead of time...

No, predictions are only qworth something if you are willing to put money behind them.

many large enterprises in this space in the USA is operating like Korean companies have for a good 80 years now. They call it 'chaebol'. Chaebol arose from family enterprises, but it really doesn't matter who or what are the controlling interests

cross shareholdings, agreements to buy, technology access, barriers to entry, assimilation of smaller upstarts into the chaebol, bonds, lending....

They’ll pop and maybe drop, but Anthropic is the Microsoft of the AI age. This is a generational company that will define the future in so many ways.
let's hope not and instead root for a more evenly distributed AI future with open weight and fully open models, running on local hardware with privacy.
I think there's a future for open weight models. To continue a half-assed analogy, Linux has thrived whilst Microsoft has captured the value. They're co-existed.

Fully open models will be more hobbyist, I expect. Open weights will continue as a strategy but probably not as prominently as today.

Why not?
Because open weight models are a business and political strategy without a strong business model.

Training is very expensive and open weight models don’t generate revenue comparable to frontier labs.

The reasons they exist are mostly “commodify your complement” and (in my, apparently controversial opinion) the Chinese govt trying to take down the American economy and undermine the US frontier labs in order to compete in the AI race.

People are so accustomed to free software (that they by and large never contribute anything to) that they get confused about why we can't just have an economy of it.
> but Anthropic is the Microsoft of the AI age

Possible, as long as everything goes well for them. Their debt deals don't care how much of a transformational company they are. If they get less demand than they expect, or slower demand than they expect, they could end up bankrupt.

I guess that makes China a generational country that will define the future in so many ways
China has existed longer than a generation.

It has already defined the future in many ways.

So long as their VPNs to Singapore stay working.
How can charge a premium over competition if they catch up? Branding, smoke and mirrors won’t last long in llm industry. Big corporation switch AI at the blink of an eye. And are already looking more into self hosting.
Well, was anyone else capable of building an operating system other than Microsoft?

The answer is that business is not just technology.

On the technical side, it's also important to see beyond the current market, what's currently possible, and broaden to - ok, many innovations are coming in AI, who is most well-positioned to be the centre of that?

Anthropic is not like Microsoft. You can’t switch os in one minute. And how many did something similar to Claude? Hundred
Right now, you’re not entirely wrong. In five years? No chance.
Notionally.

The faith/hype vs objective data doesn’t support the valuation.

It is entirely dependent on having faith that this will work as expected.

Or it could be Netscape...

Google could "Internet Explorer" Anthropic to death with a superintelligent ad stuffed free Gemini. shivers

> Or it could be Netscape...

> Google could "Internet Explorer" Anthropic to death with a superintelligent ad stuffed free Gemini. shivers

Hell, Microsoft could do the same with a cheap inferencing model; if people simply use copilot because it's already embedded into all of their applications because it's embedded into the OS (Office, Windows, anything that runs on Windows), only a few dedicated people will continue using the expensive paid-for options.

>This is a generational company that will define the future in so many ways.

You can say the same about the Company of Scotland.

If their business case is true, it has withing itself the seeds of their own death. I can ask a super smart self improving model to create a ...super smart self improving model ,-)

If their bussiness case is not true, we will be paying for this bubble in higher interest rates, for the next 25 years.

Now tell me: Which scenario do you think is more likely until 2030?

A - We cure cancer, create self improving models of hyper intelligence, that will solve Fusion and bring us to the stars, create a CPU, an OS and a 3D printed Car and deliver it to you within the same afternoon, solve Earth warming issues...

OR

B - This biggest credit bubble in history, will put the sovereign debit of the US and 30 year Treasury to 8%, force the US government to default, cut spending, increase taxes and bring about a recession that will take 25 years to recover from?

Microsoft left their competitors behind pretty quickly didn't they? They had no serious competition in desktop OS and office software for many years. Anthropic has a near equal in OpenAI and compared to Google they seem to have some advantages and some disadvantages to overcome. That's not even mentioning China models.
> This is a generational company that will define the future in so many ways.

Maybe. Or maybe not.

It’s already a generational company - $1T at year 5 is no joke.

It has already defined the future in some ways. Claude code has already part determined the future of software engineering, for example.

I am reminded of the infamous "What were you thinking?" comments from Scott McNealy. I know it's practically a cliche, and you can summon its ghost by saying "bubble" three times in a dark room lit only by a candle.

And yet.

Investors regularly place risky bets. This one if it becomes a hit is worth every risk I guess.
good chance those financial commitments will be worth more then they cost as the hardware crunch continues
Well... I guess tackling technological innovation that has a potential to transform human civilization on the most basic level does that for you. It's not acompany that releases 15th overpriced gadget to the increasingly poor populace.
but company that releases 15th overpriced gadget already transformed human civilization. Im pretty sure anthropic will have it's time and become a company releasing 15th overpriced model to the increasingly poor populace. Capitalism is cyclic: expansion, peak, contraction, trough
All of what you said both pre and post ipo also happened to Facebook.
Facebook had massive future liabilities at IPO?
Facebook didn't really have a "there's a non-zero chance that might replace a whole lot of industries" moat.
loading story #50557424
loading story #50557367
I am blown away that a company spun off from OpenAI is now worth more than OpenAI and is going IPO before OpenAI.
Turns out...in an economy hollowed out by 15 years of governments inflating asset values via money printing, 0% rates, QE & deficit spending; selling to consumers instead of asset holders (businesses) is a losing strategy.

Anthropic picked the better market at least for the short term. Long term I wouldn't count OpenAI out just yet. The consumer market tends to compound into winner take all eventually given the network effects of advertising marketplaces, even more so than B2B.

If that's true then wouldn't Google will eventually beat both of them?
Yes I think Google will outlast them both because they can just put their product into everything people use and cut the other guys out.
Google is running on search Ads revenue, it will collapse if that stream collapses, because company is too overloaded by bloat (200k employees vs 5k in Anthropic).
I'm convinced Google is a much better investment and would like to hear an opposite argument if anyone here has one.

I don't think there is a lot of difference between the "sota" models long term. Even if we said that say Anthropic is ahead, Google will be able to do the same in months.

But if it turns out that open weight models are the future, OAI and ANT are fucked, but Google has massive infra to host those and make money anyway. It also has the products and business connections to integrate AI into users workflows. And lets not forget, it earns massive amounts of money even without AI, so there is a higher price floor even if an AI bubble pops.

And even if it turns out that Anthropic takes over the world, Google still has a stake in them. (Although so will everyone else though SP500 by that point.)

I don't understand why would you invest in Anthropic IPO instead of GOOG, but as I said, would be interested to hear opposing thoughts.

I guess the opposite argument would be that AI success will undercut their search offering.
Sure, but in that case the Gemini arm of the business (or whatever they rename it to the next time) carries the torch forward. Unless there is no money to be made in AI and everyone loses, but then again it would still be better than investing in Anthropic.
> Sure, but in that case the Gemini arm of the business (or whatever they rename it to the next time) carries the torch forward. Unless there is no money to be made in AI

the question is if its enough money to replace 300B of Ads revenue.

If it isn't, are they in worse position than Anthropic?
Biggest Anthropic risk is that they have those many B hardware commitment.

Otherwise, they have some solid foundation model, which is cheap to post-train and improve, solid market share, they can milk this part for some time and be on the top.

Google is extremely vulnerable against paradigm shift if they lose Search Ads revenue, they have too much bloat of unnecessary people and systems complexity, and easily can collapse.

> I'm convinced Google is a much better investment and would like to hear an opposite argument if anyone here has one.

I have an argument, but it's mostly snark.

Like, Google are definitely in the best position here, but they've messed up so many times when it comes to product management/enterprise adoption that I am confident in their ability to snatch defeat from the jaws of victory.

> Like, Google are definitely in the best position here, but they've messed up so many times when it comes to product management/enterprise adoption that I am confident in their ability to snatch defeat from the jaws of victory.

Yeah, but what about Microsoft? They're the real threat here: they can get or train an open-weight model, embed it into Windows at very little cost to the end-user, and then the end-user will have AI in every app that they run, regardless of whether that app has AI connectors already or not.

The entity holding the business relationship already has the most clout, and while Claude takeup was good, it has nowhere near the same numbers as Google enterprise, which in turn is nowhere near Microsoft.

In terms of market-cap, these are all trillion dollar companies. In terms of enterprise agreements, if you were to draw a line-chart for enterprise volume for each of those three companies, MS would be the very top of that chart while Google would be somewhere at the bottom. Anthropic wouldn't even show up; their line would be fractions of a pixel tall, at its peak.

I 100% agree with that snark and think Google could be so much better were it not for internal politics, perverse incentives and inability to focus and follow through with products.

Still that being said, I bet on them rather than against them.

We spend 30k a month on Google services and just suffered an 18 hour outage for what our rep said to be an abuse "false positive".

So no. We will be leaving Google, and Google will not catch up.

While I sympathize with what happened to you, the argument "X happened to us therefore we will leave Google and Google will stay behind" is hardly sound. I haven't seen much better support from Anthropic or OpenAI.
The problem is Google has been famous for this for most of their services and this is their reputation. Many people have sworn they won't depend on GCP or other things.
Have you tried to get a hold of an Anthropic GTM/Salesperson to pay them to use their enterprise products? We ended up having to resort to messaging them on linkedin.
Google treats both its enterprise customers and retail customers with equal contempt.
You also need to have a working product. Google seems to do anything except creating one.
Heavy AI usage seems to be starting with work. And my usage of Claude for work has had two effects.

1. I use Claude for personal

2. I recommend Claude to others who are slower adopters

I’m not sure how your post relates to the comment your replying to?
It‘s a counterargument to the parent posrs theory that OpenAI is in a strong position because or their consumer focus.
loading story #50557051
> Anthropic picked the better market at least for the short term.

What market is that? How is it distinguished from OpenAI or any other LLM company?

I think the implication was Anthropic has gone after the B2B market (with Claude Code and Cowork), where as OpenAI has gone after the consumer market (with ChatGPT). I'm not sure I totally agree to be honest; Claude Code was a happy (and fantastic) accident if the stories are to be believed.
> I'm not sure I totally agree to be honest;

Even before Claude Code, Anthropic were way more focused on B2B (source: a former company started using them in late 2023, based on our internal evaluations).

Anthropic is much more focused on B2B while OpenAI feels more B2C. Both sell to both markets, but the focus feels different and I believe the resulting customer base is quite different.
Capital is more worthless than ever and the society is relatively poorest at the same time. Interesting times.
This post makes me realize that, despite being an enthusiastic Anthropic customer for near 2 years now, I actually don't know any of this historical context.

Down the rabbithole we go. Oh clauuuuude! :D

Helps not being commanded by, even by SV CEO standards, a psychopath.
Being commanded by a doomsday cult is better?!
Apparently yes.
They have a better model for the long term, than the Heavens Gate consultancy/doomsday cult.
Making us all pets of their machine-god is even more dystopian.
Where's the growth? Anyone really think this will hit 5T? 10T? I can't see that happening.
What are the odds you think they hit RSI?

10%? 1%?

Or do you think RSI won't be as significant as AI researchers think it will be?

Valuations are difficult for this. It's got high chance of being a bubble, but nonzero chance of being something unique across human history.

Is $1T now going to be the norm for being a Unicorn as a startup?
It's a bubble.

I am not saying AI isn't useful. I am not saying these companies don't deserve any valuation.

But it's a bubble, because investors want to get rich fast. The CEOs want to get rich fast.

What they don't think about is: When the bubble bursts, it will be difficult for many, many people. Maybe including them. Hopefully including them.

Because karma is a bitch

the valuations for these random SV "AI" companies these days is just eye watering. It really does seem like theres so much money concentrated at the top that they can throw vast amounts at the wall to see what sticks. Anthropic and Open AI are generational companies but even their values seem extremely inflated. SV has mastered the art of juicing companies to the gill before releasing them into the wild and they've done it at record speed for these two.
It’s not a bubble, enterprise is just slower to adapt than expected
loading story #50557560
loading story #50557822
loading story #50557904
And people buying it, expecting it to go up from there, heh.
There is a real scenario, whose probability I cannot accurately assess, in which the first company to build "true" superintelligence will essentially capture the bulk of the entire US economy -- and that it will do so faster than any government can meaningfully intervene. This superintelligence is not a product sold in a chatbot, but rather the internal capability represented by their most advanced unreleased models, mostly without alignment, running on the world's largest GPU clusters, coordinating millions of copies. In this scenario, they don't capture the economy simply by selling tokens, but by owning the internal capability to create (or replicate) any technology, product, or digital business on-demand faster and more cheaply than any other organization on Earth. Selling chatbot tokens would just be the "bootloader" to a future phase in which Anthropic simply enters into and takes over most markets faster than anyone else can do anything about it.

If possible to do, this is a vastly superior strategy to selling tokens, since token consumers can swap out token vendors. But if you simply recreate their entire product and technology lineups from scratch, faster and cheaper, and capture all their customers with superior products and prices (at least at first), then you own the market rather than selling tokens into it amongst other token vendors.

None of this is guaranteed to happen. But I do not see a way to rule it out as a possibility.

This is the myopia of tech people.. they exaggerate the weight of digital. AI could very well be able to replace any digital business. But purely digital businesses are such an unimportant sector of the economy. All of the important sectors use digital infra at some point, but none are purely digital.
Those businesses will simply buy tokens, as they already are. Their strategy will increasingly be steered by AI under the control of Anthropic. Their internal operations will no longer function without Claude. They'll shed staff and optimize their organizations around the extraordinary savings offered by AI. Their physical operations will gradually be automated under the control of robotics powered by Anthropic. Those who don't do this will go bankrupt in the face of competition from those who do. In some cases, perhaps many, to accelerate this they'll simply get bought out, then forced through the above process faster than it would organically occur, either by Anthropic itself or a new breed of private equity firms that specialize precisely in this kind of operation. The process will be slower at first, but not nearly as slow as you think, and nothing you mentioned stops it.
Are the designs for robots digital? If they are, they might become really good, right...?
Or just another instance of liberatrian nerd Dunning-Krueger. "Faster than any government can meaningfully intervene", c'mon - if really it is that kind of force majeure thing, they will just say "screw the free market". As if the government is bound to uphold the sacred rules of neo-liberal economics.

Same as Goedel thinking that a logical loophole in the constitution is something that could turn America into a dictatorship legally. Guess what? In practice, somebody wanting to instate a dictatorship probably won't give a crap about some subtle logical fallacy. Same with the Keynesian economics that Americans perceive as law.

You seem to think I'm invoking some kind of obscure academic theory. I'm not. This is actually the most basic competitive process imaginable playing out, and that's all it is. That's exactly why it's actually plausible.
{"dead":true,"deleted":true,"id":50564313,"parent":50557003,"time":1795113825,"type":"comment"}
The bet is that sooner or later, even the remaining meatspace holdouts will be effectively digital and thus controllable by AI.

A lot of manufacturing is already there - that's a part of why China undercuts Western manufacturers, a lot of cheap land and subsidies means you can be far cheaper than Western companies already, even more if you automate what you can. Farming, at least anything with growing plants, is also barely requiring humans any more - precision farming with RTK is absolutely insane. And what remains is trades (where, again, automation like prefabrication is a threat), parts of healthcare and education.

We will own nothing, rent everything and fight for the scraps Sam Altman hands to us.

Or we go and sabotage a couple of data centres and be done with it.
>Farming, at least anything with growing plants, is also barely requiring humans any more - precision farming with RTK is absolutely insane

In your dreams maybe, but not in actual real life.

And if you want an actual Butlerian jihad to happen in real life, then yeah, please put out of work everyone associated with farming on the planet within a few years. You describe the golden path to full-on civil war in 3/4 of the world's countries, with either AI being destroyed or an automated massacre _à la_ Terminator.

loading story #50558397
loading story #50564445
> purely digital businesses are such an unimportant sector of the economy

what? facebook? netflix? google?

selling ads is split between internet and TV + some legacy media (like it or not) + maaaaybe some outdoor

computer games are bigger than movies?

You might have a point with computer games. But if you take facebook on the other hand, it would be worthless without people using it. An AI would just create the website. For most businesses, there are non-digital elements which play an essential role, and that remains true even with AI.
In such a world, do shares in that entity help you out somehow.
Exactly! Also in such a world why would they IPO and sell so many shares (in particular if SI is imminent)? Better keep the company private and capture all the economic value created
Because they need money now and already have just about maxed out their debt load?
No idea, truthfully.
Buf after the phase...

(Stocks held during wW1 and WW2 did well afterwards)

>they don't capture the economy simply by selling tokens, but by owning the internal capability to create (or replicate) any technology, product, or digital business..

No, what that AI will do is that it will create an artificial reality (because people will perceive the world through the words of the AI) that caters to the highest bidder.

Expect this to happen in the next 20 years.

Nah. Farmers markets will just get more crowded and full of homemade items. People will move to using local currencies or Bitcoin.

A good fraction of people are smart when the system breaks. Check out the history of the Fall of the Roman Empire. And if Dan Carlin is right, become buds with your local hells angels.

Collapse of existing currency is a reasonable prediction for what might come next in this scenario, if that is how you intended your comment to function.
Love Hardcore History but haven't caught up on it in a while... What's the Hell's Angels thing?
With the reorganizing of Europe following Rome's decline, Germanic tribe leaders were handlebar-mustached strongmen. Dan Carlin analogizes them as modern Hells Angels members in looks and behavior.
Except Hells Angels are senior citizens. It’s not 1969 anymore.
Outlaw tricycle clubs are still a thing, they're still patching in, dealing drugs, and hiring mercenaries to take out rivals.

eg: https://www.abc.net.au/news/2026-02-20/david-pye-guilty-orch...

Well it says that guy is 43. That’s also Australia, that’s not how it is in the US.
> Well it says that guy is 43

and not a senior citizen, and in a gang with younger and older members.

> That’s also Australia, that’s not how it is in the US.

Gang intelligence units here in Australia say that these clubs in Australia have affiliations with gangs around the world, including the the United States.

Gangs in the USofA agree:

  There are more than 300 active OMGs within the United States, ranging in size from single chapters with five or six members to hundreds of chapters with thousands of members worldwide.  The Hells Angels, Mongols, Bandidos, Outlaws, and Sons of Silence pose a serious national domestic threat and conduct the majority of criminal activity linked to OMGs, especially activity relating to drug-trafficking and, more specifically, to cross-border drug smuggling. Because of their transnational scope, these OMGs are able to coordinate drug smuggling operations in partnership with major international drug-trafficking organizations (DTOs).
DoJ (USofA chapter, Nov 2023) - https://www.justice.gov/criminal-ocgs/gallery/outlaw-motorcy...
I’ve lived all over the US and I can tell you this is false. The only people riding around on motorcycles here are boomers and burnt out veterans.
loading story #50557193
Farmers’ markets are for rich people nostalgic for an imagined past, when rosy-cheeked, healthy maidens milked cows barehanded, with the kind of smiles only 21st-century orthodontics can deliver.
>>with the kind of smiles only 21st-century orthodontics can deliver.

love this pat of the sentence :-D

well according to mastication theorists, dental spacing/crowding was naturally in a healthier state due to increased chewing.

The fair, pink cheeked lady image is the one to be more unlikely, disfigured by cow pox scars, parasites and herpes.

In today's environment, sure. They are institutions that can expand to fill a need should it arise.

The Filipino municipal bazaars show what they could turn into when transaction costs go too high, as well as when comms and travel are difficult. They can be the lifeblood of a community.

We in the US/Europe are a little spoiled with accessibility of retail goods.

I think your orthodontics quip is inaccurate.

The dental maladies afflicting us are a product of soft, highly processed sugary food.

Lmao. If that happens, farmers are guarding their crops with their lives and the people with the most guns will be the ones in power. Your hippie fantasy won't happen, sadly.
Chaos would certainly be the first stanza in the chorus of calamity and recovery. But it normalizes -- quicker than you might think, I'll add.

The people with the most guns also sleep, and communities have long memories. There are good reasons "might makes right" doesn't scale or really hold any form of cohesion beyond a band.

> Check out the history of the Fall of the Roman Empire

Hundreds of years without sanitation?

At least they dropped the shared wiping sponge on a stick!
I don't see it working out. Before you could get any real balls rolling in the physical world competitors will be there with their own GAI, assuming we are anywhere near actually building it to start with which I seriously doubt.

And even if they are first, why assume the first is going to be the best? Maybe 6 months later a different AI model will come out that stomps all over the first one. Maybe a competitors isn't any smarter but they can do it with half the power and can scale their intelligence twice as fast, so even being later they still end up in the lead.

{"deleted":true,"id":50555790,"parent":50555759,"time":1795059525,"type":"comment"}
data moat; evidence is that we are not seeing it; and second mover advantage isn't there if they are AI-fying traditional industries.
If someone makes (super) AGI which improves recursively, open AGI will also be a thing shortly after. Not to mention all the commercial competitors (OpenAI, Google, Meta, X, etc.)

Microsoft still dominates the office tooling, and many will thus get their AGI via copilot and similar.

> owning the internal capability to create (or replicate) any technology, product, or digital business on-demand faster and more cheaply than any other organization on Earth

In this scenario, capitalism is by definition over, and market valuation becomes a meaningless concept.

I'm not sure how so many people betting on this scenario don't see that fundamental flaw: if one company can replace the whole economy, then there is no economy. In the "good" outcome the whole economy gets replaced wholesale by some flavour of "luxury space communism", and in pretty much any other outcome we end up with war/famine/etc as the system breaks down

Yeh and in this situation the people with real access to the models, who would basically control the entire world at that point, will definitely cut you in because of the piece of paper you have with a claim to 0.1% of the company /sarcasm. I mean you're talking about a capability bigger than the government, who is gonna enforce your ownership?? If you buy into their IPO you're just helping them get to this point faster.
I actually agree with you. But humans are greedy. Really, this entire situation is simply the purest possible expression of human greed.
how ywould your putative superintelligence come up with the supply chains, raw materials and presumably robots to do all this seemingly overnight? or are you only talking about digital goods and services?

even so, what about network effects? how do they lure users away from TikTok and Instagram? AI-generated clips? or say Anthropic recreates Amazon. that's a software company, right? so how do they get all the warehouses, cloud infrastructure (they will need entirely different kinds of datacenters and hardware), relationships with third-party merchants, etc.? or say SpaceX - is your superintelligence going to be setting up massive autonomous factories and buying thousands of tons of aerospace materials and billions of dollars of machinery to make that happen? where's it get the money?

and even if this somehow does manage to accomplish all that, how politically popular will it be? what prevents the US government, lobbied by.. the majority of corporations, and hounded by constituents, from invoking the Sherman Act and busting this trust like AT&T or Carnegie Steel?

In terms of supply chains, imagine a scenario where Amazon undercuts all retail stores, taking absolute control over every consumer need.

They already have solid control over logistics and inventory management.

It’s not a stretch that they then capture the means to control primary inputs and manufacturing.

Throw in an overly enthusiastic AI to continually optimise the supply chain, continually optimising a country’s supply of food and commodities, and we have all the ingredients for the classic sci-fi dystopia of a computer controlled society.

And the masses in this vision are just brain dead consumers with no agency? Market über alles?
Why do you expect your shares to be worth anything in that scenario? Note that their own timeline prediction involves a “pivotal act” in which their AI seizes the means of production.
I don't really expect anything. If that scenario actually happens, all bets are off.
> entire us economy

> But I do not see a way to rule it out as a possibility.

Your argument is unfalsifiable. Anything can happen by this logic and they could be valued whatever trillion $, like, maybe they’ll invent a way to travel through the multiverse. Maybe they’ll create the star trek replicator thing.

No. While the scenario is speculative in the sense that it is not guaranteed to occur, it is not speculative in the sense of being science fiction. It is a very mundane story of competitive capture of enough existing segments of the economy to gain negotiating leverage that surpasses a threshold. That's really it. It's actually quite boring and doesn't involve anything like what you're suggesting even remotely at all.

It is certainly not unfalsifiable. There are other scenarios in which this does not happen. For instance, the most obvious is that other organizations develop a similar capability at roughly the same point in time. This results in a "multipolar" economy that is mostly controlled by a small set of distinct owners of this kind of capability. The multipolarity prevents total capture and thereby prevents the rest of the scenario. Another obvious scenario is one in which the government acts soon enough to short-circuit this process before one organization gains a sufficiently decisive advantage that government intervention is no longer efficient enough to be effective. Yet another obvious scenario is one where a short-term financial crash deprives Anthropic of enough capital that this strategy is paused or slowed down enough to allow it to be countered. And yet another obvoius scenario is one where the scaling laws simply stop scaling or the amount of compute needed to scale them to this point cannot be physically obtained, thus allowing time for the strategy to be countered.

All of these alternatives would come with clear data points early on which would allow one to distinguish them from the original scenario I described. Right now, though, the data is ambiguous, and some data points lean slightly in favor of what I described, but not nearly enough to make any guarantees whatsoever.

Otherwise known as the "Once we create Skynet, of course it will do what we want it to" business strategy. (For the record, I think the chance of LLMs leading to any such thing is zero.)
Lol, dystopian fiction has such strong roots in our culture.

Why is it that Anthropic gains so much from this perceived image, but then needs to public money to turn it into reality? If they were at all close to any of that, surely their internal AI can figure out how to self-finance the company asap. Or maybe that's exactly what it did and it told it to go public. Company selling shovels adage.

I think markets are way more complicated and capturing them isn't going to happen the way you're suggesting: dollars, and the public, don't really like copypcats. You would need to understand the business, the product, and manipulate people at an unprecedented level. Anthropic selling tokens as their business, means that all the way up to this "super intelligence", means that they've enabled everyone's defenses and raised the threshold for this dystopian hostile taker, i.e., it won't happen.

And it will go up, until the AI hype cycle starts evaluating returns. Which could be 2027 or 2028 or 2030.
The picks and shovels are down 40% in the last two months. You can buy them at a forward P/E of less than 3 if you think earnings hold through 2029. Which is another way of saying that the market doesn't expect them to. There is a staggering amount of money being bet on "it falls apart soon and it falls apart hard."

https://seekingalpha.com/symbol/SNDK/valuation/price-earning...

That's a temporary political blip. Democrats won against data centres but in the next election the money will win over both parties.
It has nothing to do with democrats, that’s just the vacuous meme fed to retail investors. The credit market is tightening and it’s getting harder to borrow money for AI. Coreweave’s unsecured bonds trade at around 15% interest rates. Nvidia’s new CEO doesn’t believe in borrowing money to the customers. The music has stopped.
Yeah, people wanting the bubble to pop haven't been paying attention. It's been popping with the weakest entrants first - the neoclouds. ANT originally wanted to go out at a 2T valuation and had to pull back to where it is now.
Or, in the exceedingly long (in AI-land terms) timeframe you mention, they could rocket to the moon and then crash and go bankrupt because everyone uses them to bits, looks for another alternative, and local models on commodity hardware get to a point in 3 years where nobody wants to spend on tokens any more.
Models that can be ran on (high end) consumer hardware are on par with SoTA from <1 year ago. It just doesn't seem smart to buy Anthropic stock right now.
This is exactly my thinking. Hardware makers want customers too, and 3 years is a looooonng time. Only 6 years ago Anthropic was founded. Something else is being founded today that might look amazing in 3 years.
Isn't it more companies buying it rather than people?

I could imagine a company owning a substantial share could do it for other reasons other than expecting Anthropic valuation to go up.

Anthropic warns people in their IPO paperwork that they might never make a profit. Definitely not a buy for me.
Don’t basically all companies have to put that out as a disclaimer?
(In the small window of present-time, where companies IPO before they are profitable...)

Yes, that is a common disclaimer in the S-1 risks assessment.

SpaceX did this. Palantir did this. I'm not saying Anthropic is a buy, but this is essentially boilerplate.
The smart money was already in Anthropic at the series A, B, C stages in the private markets. (Which is where the quick money is made)

Are you (the IPO buyers) really expecting this company to do another 100x from its inflated valuation even after the lock up period expires?

AGI will be achieved in 15 hours and you will become their exit liquidity during the IPO fan fare.

Exactly. This is for suckers so the smart money can cash out.
They are only cashing out like $2B. Non-founder directors, employees etc can also sell 10% of their shares immediately, but most of the rest is locked up for 180 days.
At these valuations, 10% is enough for most of them to retire quite comfortably.
The employees are also the suckers in this scenario.
These lockout periods are such a joke. It sounds like we're talking about meme coins.
Yeah 6 months is really not a long time...
The suckers are going to be pension funds just like with SpaceX. The sheer size will mean OAI and Anthropic are fast tracked into indices and thus tracking ETFs that are among the most popular investment vehicles as they are low-fee, which means an immediate influx in capital followed by a permanent money spigot from pension savings.

And when it all crashes down, there will be fire sales and the vultures that are flush with cash from the exit will buy up what remains at ridiculous discounts.

We shall own nothing and rent everything, and Big Tech is obviously colluding with Big Finance to usher in neo-feudalism.

Don’t underestimate just how many suckers we’re talking here, won’t be surprised if Anthropic 10xs. Anthropic will use bitcoin style valuations.

And if they dont, then we’re all screwed

10x is unlikely, but the IPO is only ~5% of shares, and then they will release twice that fairly soon after. So it's VERY likely to see some rise early on, but not likely to be a sustained rise in pricing as the amount of shares available continues to increase significantly.
Who's "we"? IMO, if Anthropomorphic does 10x, then knowledge workers are screwed.
If it doesn't raise enough for the smart money to cash out, it stops looking so smart.
Something something Tesla will be a 30T company.
Based on a 6% float of shares. Math is doing a lot of work supporting this market cap valuation.
I am blown away by how VCs pump up some companies to eventually get access to your 401Ks without any questions neither from government nor from public.

I will be double blown away when bubble pops and some companies go bankrupt or bail out by the government, who already paid VCs with your money.

Winners: VCs

Index is proportional to free float, so Anthropic will be treated as $100B company by index, not $1.5T company.
loading story #50559170
Wait until it explodes to 3-4b shortly after IPO, then... I think that despite the heavy selloff in the AI trade this las month we are not yet past the peak frenzy of the cycle.
I wonder on how they're planning to impose that valuation on the people of Earth.