Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Apple Will 'Watch Everything Burn' When AI Bubble Bursts - Ed Zitron
- Thread starter MacRumors
- Start date
- Sort by reaction score
Apple's stock is down because it always does so whenever Apple posts revenue despite how well they're doing. The AI related stocks have been getting batterred for the most part for couple of weeks. SpaceX is down almost 50% from its high a month ago. The other big AI companies OpenAI and Anthropic you can't tell since they're not public IPOs at the moment. This post of yours is just ridiculous
pankajdoharey
macrumors 6502a
Who is this guy? Does he even understand that OpenAI is the fastest-growing app in the history of the internet? In three years, it reached revenue of $13.6 billion. If its expenses are $20 billion, that is a deficit of $7 billion. That is nothing for a company growing at this pace.
Google took five years to make its first billion. Amazon took five years to reach $1 billion. Facebook took six years. Tesla took ten years. Twitter never reached a billion in revenue until after it became a public company. The moral of the story is that we have never seen companies grow this fast before. No technology has proliferated as quickly as machine intelligence.
The hyperscaler buildouts are not being built as fast as consumer demand, which is why there are arbitrary quota restrictions. The bottleneck is actually Nvidia’s hardware, Power Production, land allocation, Opposition by states who dont want data centers, political stupidity, clearance issues, etc .... And Nvidia is making a killing. It is easily priced at 20 times higher than what it costs to make. Even RAM makers are selling a trivial chip at a four-times markup.
That said, there is a stock market bubble, not an AI bubble. The technology is real, and people are using it. Just look at the downloads of open models, and you will see that this technology is being used crazily. Closed models are even more capable. The growth trajectory of these companies is exponential. This will only stop when demand declines, which is not happening.
This is the second Industrial Revolution. The first Industrial Revolution replaced human labor; this one will replace human intelligence. It is far more impactful than the Industrial Revolution, because the first produced machines to automate manual labor, while this replaces intelligence itself. If some numbskull can’t understand the impact of replacing intelligence, it is because they don’t have any.
Google took five years to make its first billion. Amazon took five years to reach $1 billion. Facebook took six years. Tesla took ten years. Twitter never reached a billion in revenue until after it became a public company. The moral of the story is that we have never seen companies grow this fast before. No technology has proliferated as quickly as machine intelligence.
The hyperscaler buildouts are not being built as fast as consumer demand, which is why there are arbitrary quota restrictions. The bottleneck is actually Nvidia’s hardware, Power Production, land allocation, Opposition by states who dont want data centers, political stupidity, clearance issues, etc .... And Nvidia is making a killing. It is easily priced at 20 times higher than what it costs to make. Even RAM makers are selling a trivial chip at a four-times markup.
That said, there is a stock market bubble, not an AI bubble. The technology is real, and people are using it. Just look at the downloads of open models, and you will see that this technology is being used crazily. Closed models are even more capable. The growth trajectory of these companies is exponential. This will only stop when demand declines, which is not happening.
This is the second Industrial Revolution. The first Industrial Revolution replaced human labor; this one will replace human intelligence. It is far more impactful than the Industrial Revolution, because the first produced machines to automate manual labor, while this replaces intelligence itself. If some numbskull can’t understand the impact of replacing intelligence, it is because they don’t have any.
IvyKing
macrumors 6502a
The question is whether the trillion dollar investments for hardware will pay off. Or another way to look at it is how much are the customers willing to pay the likes of OpenAI?
In the long run, yes it pays off. People criticise “AI circular financing” seem to be ignorant about how financing and investment works. Nvidia is taking its fresh record breaking revenues and ploughing it back into the industry which is creating growth and jobs.
Would they prefer that Nvidia just sit on a mountain of cash and do nothing with it?
People like this Zitron fella are not different from those who argued agains tthe cost of building cities, roads, railways and the space program. They are always wrong. Most infrastructure humanity builds is super expensive but in the end everyone benefits.
We hate stinky cities and noisy railways. Urbanisation, tech and science development is better than beating each other with rocks in the jungle while believing in sky wizards. The costs of development outweigh zero growth or backwardness.
AI has been under development for almost the entire history of computing. It’s not new. Every cent the computing industry has earned since the middle of the last century has been investing towards this ultimate vision of making computers intelligent so that work becomes easier, medical and technological development becomes faster, discoveries are made faster, and we build parallel virtual economies and space based economies with the help of AI.
If there is no AI future then there is zero human progress from here. We can’t have future progress if people just want snappier browsing and smoother window resizing and better GTA graphics. That’s not going to pay the bills for generations or grow the pension pots.
Last edited:
Everybody also has the "savvy" nephew who was telling you to put all your money in NFTs, SafeMoon, or fidget spinners.
Companies I've been keeping an eye on regarding the bubble thing is Tesla/SpaceX, which have lost a ton of progress in their stock, particularly SpaceX, which has lost half of their high a month of so ago which is not a good thing for them. Granted, with any of the Musk companies, its hard to tell if its part of the bubble bursting, or Musk's antics poisoning his own companies. Granted, Tesla's stock is insanely over priced compared to what they actually produce. Another is Oracle, which is interesting to see considering how much the Ellisons have tied in to the pending WB/Paramount merger and how that could fail if Oracle takes a big enough hit. Zitron is not wrong. There is a bubble and it will burst. I thought it would happen last year but its taking longer than I thought. But something is about to break. The bubble is just part of the house of cards and something is going to knock the whole thing down. The return of tariffs, the continuing war, the combination of the 2 causing the prices to rise on everything. If people can't afford things, they're not going to buy things. If they can't buy things, companies don't get revenue. Companies don't get revenue, investors lose out. The investors lose out, they can't continue to fund the AI startups and so on. Or it could go in the other direction too. One of these AI companies fail that triggers a panic with the rest of them which affects the stocks of everything else, coupled with the issues I mentioned earlier and we're in a massive recession. It's not an if we're going to be in a recession, its a matter of when. We're overdue.
I think of the downturn during COVID and the stimulus that helped pull us out. But, I'm old enough to remember the Dot com bubble bursting and 9/11 happeneing shortly afterward and the downturn, and we had stimus and deregulation that helped pull us out of that but all of that led to the situation that created the great recession of 2008. I remember in 2006 and 2007, we'd have all these highs in stocks and people ignoring those who said its a bubble and the growth was unsustainable and because they were looking at the underlying issues like Ed was looking at. It took a year or 2, but it came down big time.
The big question is will this be a recession like 1990 that was triggered by the Saving and loan crisis and rising oil prices tied to the first gulf war? Or will this be like 2008 or bigger?
I think of the downturn during COVID and the stimulus that helped pull us out. But, I'm old enough to remember the Dot com bubble bursting and 9/11 happeneing shortly afterward and the downturn, and we had stimus and deregulation that helped pull us out of that but all of that led to the situation that created the great recession of 2008. I remember in 2006 and 2007, we'd have all these highs in stocks and people ignoring those who said its a bubble and the growth was unsustainable and because they were looking at the underlying issues like Ed was looking at. It took a year or 2, but it came down big time.
The big question is will this be a recession like 1990 that was triggered by the Saving and loan crisis and rising oil prices tied to the first gulf war? Or will this be like 2008 or bigger?
If you believe the field of deep learning is like “NFTs” you have been misled by unskilled and untalented critics (who are only critics because people pay for their newsletters) into believing deep learning is some two year old fad when it is a 70 years old scientific field. Being in denial about deep learning is science denialism. If you don’t want to be replaced by someone who knows AI skills then be smart instead of being misled by newsletter scammers. It will like be the difference between knowing how to use a computer and not knowing how to use a computer in the 80s and 90s.
If you believed a “bubble was going to burst” last year that discredits your whole thesis.
Anyone who saw the state of models last year, the rapid progress and the user growth would not even come close to believing what you did.
We are nowhere near the best of what generative AI and deep learning can do. The enormous growth that is coming in the form of physical AI is the next stage. AI enabled robotics are going to be everywhere, some running local models and some running off remote models, and some a mix of both.
If futurists in the 1950s could see that coming then nobody has an excuse not to see it coming in 2020s. A few thousand misinformed privelaged westerners sobbing about AI on the internet is not going to stop billions of users and is not going to stop Chinese and Indian start ups from trying to become number one.
The question is : be in denial and lose the financial opportunities of the decade or be active and get rich?
Just don’t listen to people who are consistently wrong in their subscription newsletter and youtube channels. They are often manipulating you on behalf of someone else and do not believe what they say.
Last edited:
Finally want to add one important note.
There is an old story about a man who was called crazy because he built a big expensive boat and filled it with animals because he feared a dangerous natural event was going to happen.
Within the next century or two it is extremely likely humanity will experience a volcanic event or asteroid event that will pose a mass extinction threat to all of us. We need the assistance of robotics, space travel and super intelligence to help us survive because we cannot do it alone. We are too slow, too physically weak, and too self centered and selfish.
AI will help us stay alive on our own planet and in the new frontier of outer space. Anyone standing in the way of that wants extinction. They want all of us to give up just because they have given up. We can’t let a minority of defeatists and science deniers stand in the way of survival of the whole species.
There is an old story about a man who was called crazy because he built a big expensive boat and filled it with animals because he feared a dangerous natural event was going to happen.
Within the next century or two it is extremely likely humanity will experience a volcanic event or asteroid event that will pose a mass extinction threat to all of us. We need the assistance of robotics, space travel and super intelligence to help us survive because we cannot do it alone. We are too slow, too physically weak, and too self centered and selfish.
AI will help us stay alive on our own planet and in the new frontier of outer space. Anyone standing in the way of that wants extinction. They want all of us to give up just because they have given up. We can’t let a minority of defeatists and science deniers stand in the way of survival of the whole species.
I agree that AI will continue to have a huge impact and will continue to develop rapidly for a long time. This does not negate the probability of a bubble bursting in the valuation of a lot of firms that are competing with each other and producing models that are cheaper than before. The bursting of the dotcom bubble did not end rapid progress in electronics, it simply brought about a major correction re the overall profitability of many individual firms involved in that field.
Exactly. I think the bubble bursting will get people's priorities straight when it comes to AI.
AI is going nowhere but there will be a crash. There is something of value in this tech but it isn’t going to lead to AGI.
I think its funny that he says "AI will help us stay alive on our own planet" At the rate we keep building data centers, all AI is going to help us do is to roast us alive
Apple keeps falling in the stock market and all AI and space related companies are back on the upward trajectory.
The moral of the story is always, you can’t stop the future. Whenever a scientific discovery, invention or technology is ready for mass delivery then nothing can stop it except for backwardness and corruption.
There has been recently the phenomena of English guys with YouTube channels and subscription newsletters who pretend they are “one of the people” who are “fighting corporations”. These people are frauds who have been groomed and trained to manipulate markets and brainwash young people to hate progress. They rake in millions of views, do TV interviews and money from subscriptions but don’t really believe a word they say. They take your money, laugh behind your back and get rich.
The moral of the story is always, you can’t stop the future. Whenever a scientific discovery, invention or technology is ready for mass delivery then nothing can stop it except for backwardness and corruption.
There has been recently the phenomena of English guys with YouTube channels and subscription newsletters who pretend they are “one of the people” who are “fighting corporations”. These people are frauds who have been groomed and trained to manipulate markets and brainwash young people to hate progress. They rake in millions of views, do TV interviews and money from subscriptions but don’t really believe a word they say. They take your money, laugh behind your back and get rich.
Yes, and humans who simply copy others' work lose copyright infringement cases. Creative humans learn from their predecessors and then create something new. This is what LLM AI can't do.
Humans are capable of observing and gaining an understanding of others' work. What gets committed to memory is that understanding. When creating something new, their work is built on that understanding. LLM AI has no understanding, but it does have an eidetic memory in the form of a massive cataloged database storing copies of others' work. When prompted to create something new, AI turns to that database, clips parts from those copies and reassembles them based on a probabilistic algorithm. This is not what human artists do.
Couldn't have said it better myself. When I see these AI supporters counter the arguement from skeptics about the theft of intellectual properties with "oh well you look at other artists too when you first learned", I'm like "Yes, we look at other artist's work and yes that does influence what we do. But what we do that AI cannot do is take into account our personal experiences, things not even related to the actual art and that influences us just as much if not more than the artists we look at. I'm an illustrator and though I was influenced by other illustrators, but I was also influenced by music, movies, and just life itself. And that personal experience is what gives our work the soul that AI generated art lacks and will always lack".
Nobody except bad artists are scared of AI. Artists of all kinds who embraced technology have always been leaders in their professions. Illustrators in particular have been very impressed by generative AI updates to Adobe Illustrator that allow them to do things that previously would be hard to imagine or take too long to do.
gleepskip
Suspended
Commercial flights cheapen the experience of travel. Cars do, too. If you want to experience authentic journeys, they need to be made by stagecoach on unpaved roads.
IvyKing
macrumors 6502a
I don't think you have any clue as to how power contracts work - the lines supplying the data centers will most certainly be metered.
thebarrkbyte
macrumors member
Oh, I agree. Never let nerds define education methods. They always throw tech at kids and it never works. Over organization and too much structure stifles children's creativity, as well.
Kids need people, strong socialization with adults to learn strongly, and with other kids to learn roles. What is some dumb computer going to do better than a trained person with a warm heart and strong passion?
A computer is a just a dumb terminal to become a narcissistic reflection of the one who owns it.
Last edited:
Has the bUbbLE poPPeD yet?
Didn’t think so. GPU and memory prices rose 5% since this thread started.
SK Hynix is building three more mega fabs but they won’t be functional for 18 months at least.
Chinese fabs are coming online but they can barely meet China demand let alone the whole planet and they do not compete with HBM4. They are still struggling to produce HBM3 in any volume.
HMB4e is even faster and the demand is not slowing down.
Quantum, Solar and Optical stocks have been popping off too not just GPU and DRAM. A whole global infrastructure is being upgraded and revamped, even the networking needs to be upgraded.
And then space based compute will come next. Chinese companies ADA Space and Zhejiang Lab, Google, Amazon and SpaceX are the main contenders at the moment.
Be part of the future and don’t listen to English influencers who write for the vile Guardian and want to keep you poor and feeling victimized.
turbineseaplane
macrumors Penryn
Do you realize that even Sam Altman has said we are in an AI bubble?
Hi, let my 64 year old self teach you 64 years of observation from repeating cycles and repeating doomsayers.
People are often wrong and sometimes people in influential positions are trying to manipulate others. Either they try to manipulate the market or they are trying to demoralize competing companies.
The future is always being built. Don’t be a loser who thinks the future isn’t being built and all tech remains static. Everything has been leading towards AI, robotics and space colonization. Computing exists for this very purpose. All these corporations haven’t been inventing technologies just so you can post a little online, watching some videos and check if Safari is snappy.
Don’t be this like famous guy who was right once and wrong 100 times.