Anthropic is profitable if you ignore how much it spends.
>miss out on the next amazon because your too dumb to learn from history and see what that doesn't matter even a tiny bit
>>110097941We have technology that can put Mathematicians out of work.AI companies are not going anywhere.
>>110098525>the next amazonAre you fucking retarded?
>>110098581AI is going to be far bigger than Amazon because it can take over every industry.Art, Medicine, Finance, Agriculture, Science etc. There's no stopping it and no limits either.
>>110098605you're buying in AFTER it already got big.
>>110098556Come back when OpenAI becomes profitable or Anthropic stops faking its revenue.
>>110097941its going to make money obsolete
>>110098623It's tiny. Of course, once it takes over everything, money probably won't matter, one way or another.
>>110098736if you think that then why even bother investing?
>>110098556Yes, all that maths money, that mathematicians were definitely earning through their maths and not by hugging the grant grif university prestige cycle...
>>110098773Yeah, good question. The success case is the end of the world as we know it, for better or worse. What's the scenario where investors are meaningfully rewarded?
The only objective improvement AI brought so far is marginally better Microsoft Sam TTS that kinda resembles Bruce Campbell.
>>110098525Amazon had an extremely obvious path to profitability. Anthropic doesn't. Also Anthropic can't stop spending absurd amounts of money because if they do they'll fall behind Microsoft and Google and be no different than free models.
>>110098781It's what you can do with those maths, idiot. 'Murrica spent lots of money building the atomic bomb but the end result was subjugating the entire world, including the country they were at war with (Japan). AI is basically the same thing on steroids. They can create new bombs, medicine, computers that nobody else on Earth can compete with.
>>110098815>They can create new bombs, medicine, computers that nobody else on Earth can compete with.So far, vibe coders are churring out tons on slop and nothing useful. And yes, the Navier Stokes thing ...
>>110098815>They can create new bombs, medicine, computers that nobody else on Earth can compete with.
>>110098825Jesus, still bringing up 2022 talking points?We went from AI that could barely write a sentence, to replacing entire companies like Adobe.
>>110098835It's a game of cat and mouse anon.Lets say OpenAI does have the technology to create a new bomb.Why in god's name would they leak it, which would cause America's enemies to panic? Think about it before you post your zinger.
>>110098845>to replacing entire companies like Adobe.Has Adobe filed for bankruptcy ?
>>110098525You know nothing about Amazon's early days.
>>110098835>>110098856Also, the atomic bomb was kept secret.Japan was just like you and denied it.>HAHA. DUMB AMERICAN WASTE MONEY ON SLOP. THEY SO STUPID. It took a second explosion to make them shut up.
>>110098885if its secret then how do you know it exists?
>>110098869Give it 6 months. They're the next Blockbuster.
>>110098788Perhaps the biggest investors get to be among those they don't round up and cull
>>110098896You're talking about the nukes? It got declassified later. Also, the Soviets had spies who were watching the Manhattan Project.
>>110097941Remember2008 2 is nearest
>>110098909It's pretty much this. Buying an Openai or Anthropic share spares you from the apocalypse.
>>110098908Maybe two more weeks?
>>110098943Don't cry when reality hits you in the face like it did in Japan. >>110098885
>>110098869The sloppers are hard at work, but for some reason, as the software gets larger and larger, more and more bugs creep in and the AI gets progressively worse at solving those bugs.Even more puzzling, real professionals don't want to switch over to buggy vibeslop that was announced last tuesday and that its own makers don't really understand.
>>110098845>All imitations of the real thingSo AI is like a supercharged Chinese factory worker But can it come up with something novel and useful?
>>110098980You forgot the part that the Adobe replacement software will be abandonware in a month.
>>110098980>Even more puzzling, real professionals don't want to switch over to buggy vibeslop that was announced last tuesday and that its own makers don't really understand.No one cares about the tranny artists who already got laid off from their jobs.The AI clones are doing fine and getting better each day.
>>110098815Real life does not work like a Captain Marvel comic, anon. Also how does any of that have anything to do with Anthropic making money?
>>110098773You could be right about (AI eats the economy) and wrong about (luxury space communism), in which case owning even one Anth share could be the difference between getting old in the UBI slums and galaxy ownership.
>>110098997>but what about AI job stealingWhat about it? Everybody hates that shit.
>>110098556>thousands of math problems>only 7 solved and confirmed the rest is all AI hallucinated "solutions"basically, buy an ad
>>110098998>Real life does not work like a Captain Marvel comic, anon.AI is solving math problems faster than humans can even have time to fully understand them.We crossed the rubicon a long time ago.>Also how does any of that have anything to do with Anthropic making money?Create a cure for cancer = sell it on the world market for trillions of dollars.Or.Create a new fighter jet that can never be shot down = sell it to the highest bidder.
>>110098992No way, vibemongos have the attention span of an omniscent god, so there's no way they will forget it next week when someone puts a Skateboard into Spiderman 2, and there's no way that plummet will have any effect on that project.Surely, the hype will continue forever and they have infinite Claude tokens. With YouTubers making "I tried the new vibe coded Adobe alternative. It didn't work" videos, how can the excitement possibly die?INVEST INVEST INVESTIt's going to the moon.
>>110097941I like how they get a little bonus to keep them rated at the stretchiest ratings possible.>Well, we can’t not call them junk without losing credibility, let’s go with the highest level of junk.>Oracle’s pretty much junk, but, maybe we can put the rating just above the junk line to keep it included in certain funds that aren’t paying attention.
>>110099019AI understands exactly zero math problems that humans didn't entirely figure out ahead of time.
>>110099018>>only 7 solved and confirmed the rest is all AI hallucinated "solutions"Feel free to show us your solution to Navier-Stokes. You had 90 years to try but you didn't do shit.
>>110099036Holy fuck, not even Tao agrees with you even anymore.Stay assblasted forever.
>>110098525Amazon didn’t IPO at an over 2 trillion Mcap, lmao.
>>110098845Except all of those are unusable garbage that zero professionals actually use and Adobe is still raking in billions annually.
>>110099050>but what about this random twitter post I foundlol>>110099041Feel free to show us yours. OpenAI's solution to that problem is literally just news bullshit material for now, it still isn't accepted or rejected.
>>110099076No one cares Dungpreet.
>>110099109That's the best you can come up with?
>>110099132>SAARS! Please give izzat.
So the business model of AI companies is essentially selling processing power (tokens) to use their powerful calculator. Am I missing something? The only path to profitability is the price of the tokens. And they need to make them way more expensive than human wages to even come close to profitability.
>>110099164LOL so that really is the best you can come up with.>Fact: nobody uses this software>UR INDIANbest you can do
>>110098845Exactly the kind of mystery meat mongoloid that would be excited over this shit. When is AI going to MAKE SOMETHING NEW? If you're just copying and stealing from good shit, why don't I just use the good shit?
>>110099181>SAARS! Do not redeem!
>>110099168You're forgetting the hype train in between your third sentence and your fourth sentence.
>>110099019Sounds like somebody never watched his Mister Rogers growing up...https://www.youtube.com/watch?v=ZdUT9nknA0I
>>110099168> Am I missing something?Yes. They have contracts with the U.S government and enterprise businesses.The tokens are just a rounding error.
>>110099186>unironic bot postNobody uses AI software.
>>110099201>SAARS! You are the bot posting!
>>110099200No, he has it right, government and enterprise businesses exclusively lose money by using AI as opposed to what they were doing before.
>>110099200But when Adobe sells their shit to enterprise business, there's just Adobe. There's nobody else in the game but Adobe. There are like 10 powerful calculators that can achieve the same thing for your average enterprise business.
>>110099208>unironic bot postNobody uses AI software.
>>110099211>as opposed to what they were doing before.AI allows them to layoff employees and double productivity.Payrolls are the biggest expense at every company.
>>110099220Three things that aren't true.
>>110099217>SAARS! I am Brahim class now.
>>110099230>unironic bot postNobody uses AI software.
>>110099229Then you've never watched the news.Mass layoffs are happening but companies are still reporting a profit?How could that be, retard? It's because using AI is cheaper than paying each employee $100,000+ a year.
>>110099239>SAARS! You are low on Izzat!
>>110099241Is the AI company profitable?
>>110099250>>110099230>>110099208>>110099186>>110099164kek bodied the snaicat
>>110099241>How could that be, retard?Because it isn't true. Most companies that do what you're describing stop making a profit. Especially consulting companies, the ones that were the first to actually go 100% AI. To do what you're describing, as in only AI matters and only what the AI says is what goes. Companies like MIcrosoft are actually only gradually shifting over to AI in comparison, and so far as they've transitioned the profits for them are terrible.Investment isn't profit, btw, but you knew that already, since you're a bot trying to defend investment.
>>110099250>unironic bot postNobody uses AI software.
>>110099253Yes, they are.But faggots will still find a way to seethe.
>>110099258*snailcat
>>110099266>SAARS! You have copied and pasted that post. You are the botting.
>>110098810>Amazon had an extremely obvious path to profitability.Totally wrong.AWS and datacentre cloud server rentals was the only profitable portion of Amazon, and it was only started 12 years after Amazon was founded.
>>110099271>on an adjusted basis>>110099277>unironic bot postNobody uses AI software.
>>110099271May I have an analysis that doesn't end with "We said so?"
>>110099259>Most companies that do what you're describing stop making a profit.Disney was profitable this year dumbass.And they still have enough money to buy out other companies.Now shut up.
>>110098525Amazon immediately started with a clear path to profitability. They didn't go "hurr durr we'll figure out later".>>110098810I think there's a point to be made as to how the massive disparity between the compute required to produce a model and that required to run it means that producing AI models is inherently unprofitable. Since the technology keeps becoming more efficient over time, we are starting to reach the point of diminishing returns where many local models are "good enough" for most non-technical use cases and efficacy is becoming more reliant on the capability of the harness rather than the model itself. There's also the issue that models are becoming more capable at writing solid code which makes selling harnesses also not a good venture. I think in the end, most of the money will be made in selling hosting for models, and the winners are going to be companies that host models but don't waste money in developing new ones.
>>110097941note that only pure plays end up asking daddy government for AI pause. meta and google and the Chinese dgaf
>>110099292>that doesn't end with "We said so?"That's every company, lol.
>>110099285lmao, no. You're trying to equate margins to "obvious profitability."You're right in that AWS is a giant money bag, but you're wrong in contradiction to the guy you're replying to. Amazon is Wal Mart, which is to say that profit margins are low but also guaranteed. There's no way with Amazon's business model that they won't make consistent money through sales of product. The reason for investing early on was to get a large chuck of that retail pie that Amazon absorbed with consistent profitability.Small margins are fine if they are guaranteed.
>>110099294The majority of Disney's revenue is determined by activities that have nothing to do with AI, despite Disney's best efforts.Your own article shows a picture of a Disney park as an example. Is the Disney park AI???
>>110099307>I'm wrong but don't think about itNo.
>>110098810>Amazon had an extremely obvious path to profitability.How old are you?
>>110099360How old are you? By the time Amazon went public, the bogus myth that libraries would compete with Amazon was long dead.
>>110099340>The majority of Disney's revenue is determined by activities that have nothing to do with AINo, you intentionally ignored the previous article where it was confirmed they laid people off >>110099241Employees = the payroll I talked about earlier. They're shedding a major expense which helps them be profitable.
>>110099340Oh, and they are heavily using AI as well.
>>110099018>>only 7 solved and confirmed the rest is all AI hallucinated "solutions"This is just retarded. They literally have lean certificateshttps://github.com/openai/mathhttps://github.com/openai/NavierStokesAndEulerObviously not all of this has been checked, but remember that only the problem statement has to be checked for most problems they posted, as they used lean to prove the theorems. Also what are you going to do when top mathematicians confirm it's all correct except maybe 3% of them?
>>110099409No I didn't, you're dancing around the fact that Disney isn't going to benefit from AI despite laying off a few hundred employees.Meanwhlie you're ignoring what I said >>110099259 here where consulting companies that ACTUALLY went in on AI, where they got rid of employees on a cost benefit analysis of AI, hence the AI economy you are being paid to post here about, actually didn't make money and are struggling or out of business.Disney is "doing the same" because they aren't as in on AI as you are pushing. They'll do worse if they keep investing.
>>110099050It sort of reaffirms his point though. The AI only knows what it was trained on. The derivative of it was more of a chance thing.
>>110098556Usecase for maths?
>>110099456>No I didn't, you're dancing around the fact that Disney isn't going to benefit from AINow you're just being annoying.Fuck you.
>>110099479Yeah just ignore the rest of the post while pretending you have a legitimate ego.That'll really convince literally anyone ever.
>>110099468>It sort of reaffirms his point though.Tao literally admits ai made massive breakthroughs. Yes, it used previous work. Ok and? Mathematicians do the exact same thing. Can we please stop moving the goalpost? AI *LITERALLY* solved the Navier Stokes millennium problem. What more proof do we need?
>>110099468>The AI only knows what it was trained on. So why didn't humans solve Navier Stokes in over 90 years then? If the answer was always there that would imply:1. Mathematicians were frauds whose jobs are clearly overpaid.2. They're incompetent and we needed computers to rescue them.Either option only justifies letting AI handle maths from now on.
>>110099494Disney has every reason to benefit from AI.The easiest example comes from their movies. They already cost millions to make. AI ensures they can shrink those team sizes and use the money for other projects.
>>110099513An AI movie on the scale of a Disney movie costs billions to make.
>>110098885>Japan was just like you and denied it.No they didn't.Not that anon, and I'm on your side in your argument, but holy shit you are a historically illiterate cur.Japan knew a weapon from the energy of nuclear fission was possible, and was trying to build their own.As were the Germans and Brits. Japan was further along than both of them.Please use something you actually KNOW something about for a metaphor.
>>110099516>An AI movie on the scale of a Disney movie costs billions to make.If youtubers can make high quality VFX at home with AI, then Disney can do the same.Get your head out your ass boy.
The models will still exist even if these companies implode right?
>>110099518Japan was in denial when the first atomic bomb dropped. There was no response in THREE DAYS, which caused America to drop the second.
>>110099524>If youtubers can make high quality VFX at home with AI, then Disney can do the same.NTA but if this were true, why would anyone watch Disney movies and not slop movies generated at home for free? If anything, Disney is fucked (same as adobe)
>>110098885Just stop posting, retard
>>110099524Are these YouTubers named some variant of Kumar or Denish?
>>110099552>NTA but if this were true, why would anyone watch Disney movies and not slop movies generated at home for free?Why do you think Disney movies have so much merchandising and licensing deals?The movies are literally toy commercials to get the people to consume their other stuff.If anything, Disney was way ahead of you. They'll happily sacrifice their movies if you go buy 10 shirts instead.
>>110099532Exactly. I don't know why so many people are so concerned about ai companies profitability. Assume they all fail: software engineering is still dead, tech work is still dead, mass layoffs due to ai are still coming, computer skills are still deprecated etc. AI models aren't going anywhere, even if the jewish companies fail (they won't btw, but let's assume it for the sake of argument), since these models are literally just files.
>>110099564Greetings Hiroo Onoda.
>>110097941I'm probably buying 20k on IPO, even though I know it's a bad idea. Despite everything, it seems undeniable that utility exists and their models are still world class.
>>110099578Not so fast! Every model will eventually turn to shit and AI will be dead forever. https://www.reddit.com/r/BetterOffline/comments/1x8gdk6/large_scale_model_collapse/?utm_source=app_first_navigation&mweb_loid=t2_2p0dwafbeo
