AI Stocks Balloon Even As Earnings Lag, Jefferies Warns (indiadispatch.com) 57
An anonymous reader shares a report: A basket of 27 large-cap AI stocks created by wealth manager and brokerage house Jefferies has surged 127% in value since ChatGPT's launch in late 2022, adding about $10 trillion in market cap. However, 2025 earnings forecasts for these companies have increased only 25% over the same period, Jefferies warned in a note to clients.
This disconnect has pushed the incremental price-to-earnings ratio for AI stocks to 73 times, suggesting investors are pricing in extremely optimistic growth expectations across the sector.
Nvidia has seen the largest gains, with its stock price up 656% since late 2022. Despite signs of overvaluation, Jefferies believes the AI bubble could keep expanding in the near term, citing strong capital expenditure plans through 2025 and ample cash reserves at major cloud providers.
This disconnect has pushed the incremental price-to-earnings ratio for AI stocks to 73 times, suggesting investors are pricing in extremely optimistic growth expectations across the sector.
Nvidia has seen the largest gains, with its stock price up 656% since late 2022. Despite signs of overvaluation, Jefferies believes the AI bubble could keep expanding in the near term, citing strong capital expenditure plans through 2025 and ample cash reserves at major cloud providers.
A who the what now? (Score:5, Funny)
This is Slashdot; we know Jeffries Tubes, but WTF is a Jeffries Basket?
Re:A who the what now? (Score:4, Funny)
WTF is a Jeffries Basket?
It's what they put on the end of a Jeffries Tube to catch the output.
You don't want that stuff just leaking all over the deck.
Have they tried remodulating the shield frequency? (Score:3)
Seems to be the "greater fool theory" at work (Score:4, Insightful)
Nobody halfway smart can still expect the current hype-AI will actually bring great profits.
Re: (Score:2)
Anyone who thinks that the average Venture Capitalist is even halfway smart is a gullible fool, just like most Venture Capitalists. I'm always amused when people suggest that Capitalism, especially the current version of it, is in any way efficient. It's actually difficult to come up with a less efficient form of economy. Of course it is arguable that the actual benefits of Capitalism outweigh the downsides, including this one. I personally don't think they do, and a more pure form of market economy without
Re: (Score:2)
Like many things with the human race, capitalism works reasonably well small-size, with small business owners, shows, self-employed craftspersons, and so on. Scale it up and it goes to hell where people have to work, work, work and still struggle, despite massive industrial productivity and some assholes get rich to an unimaginable and completely insane degree.
A bubble (Score:5, Insightful)
"* Stocks ballon even as earnings lag" is more or less the definition of a bubble for whatever matches the glob.
Literally no one saw this coming.
Re: (Score:2)
Re: (Score:2)
Oh boy. You're right, I do love it.
Can't wait for it to his slashdot and have all the EM drive believers come back in force because this time someone has successfully invented a drive/perpetual motion machine.
EM drive is a different animal (Score:1)
> The EM drive is a thinly disguised perpetual motion machine. [sig]
For many of us it was more hopeful thinking than logic, like "yes, the claimed physics is dodgy, but maybe they accidentally found a way to easily turn matter into energy by bouncing radio waves in a lopsided echo-chamber." I wouldn't purchase stocks in it, but just hoping we stumbled into our Zefram Cochrane moment. It's okay to have hope, just don't bet the farm on hope alone.
Re: (Score:2)
Yeah don't get me wrong, I would love[*] these to be true. The idea of zipping around space like all of my favourite space operas is grand.
Unfortunately, reactionless drives more efficient than a photon drive are free energy device, a.k.a. perpetual motion machines. Even though physics isn't fully understood I feel more confident that laws of thermodynamics hold than anything else.
[*] Kind of, but the consequences are dire. If you have these then anyone with an interplanetary shuttle craft is now essentiall
Re: (Score:1)
> perpetual motion machines.
One theory was that the radio waves were turning the atoms of the reflection chamber into energy, not that energy was coming out of nowhere. Matter is a very compact energy store. The hard part is harnessing it. Each gram of matter is roughly 1 Hiroshima bomb's worth. Eventually the chamber would grow lighter, but after it's on the other side of he galaxy.
Re:EM drive is a different animal [correction] (Score:1)
correction: "other side of the galaxy". (There is no Helium Galaxy that I know of.)
Re: (Score:2)
One theory was that the radio waves were turning the atoms of the reflection chamber into energy
Sure, but that's a different thing to the one I'm referring to. Specifically, thrust without reaction mass is equivalent to a perpetual motion machine. With the EM drive they made no claims as such, but it's a consequence of physics.
Basically, they claim they can generate a given thrust (Force) with a given power level, i.e. F = P*e where "e" is the hypothetical drive efficiency factor in Newtons/Watt.
But we also
Re: (Score:2)
> or an even more exotic thing that somehow unbinds matter, the a working EM drive still violates conservation of energy.
Please elaborate. You agreed matter is a compact energy store. It's not the energy of the radio waves that would be generating the thrust energy, but rather the side-effect of the radio waves converting the metal chamber's molecules into energy (matter-to-energy). You seem to be confusing the radio's power with the energy source.
I agree that with the prototypes it took a decent chunk o
Re: (Score:2)
Please elaborate.
Sure.
Imagine you have a hypothetical drive which generates 1N of thrust per Watt of power input in free space with no propellant (i.e. it's not using, say, magnetic fields of nearby planets and so using a planet as reaction mass), the mass of the device does not change.
Assume the spaceship including drive and battery powering it weighs 1Kg, and the power fed into the drive is 1W.
Initially, the drive is at rest (relative to the observer). The drive runs for 100 seconds. During this time, it
Re: (Score:2)
Considering there is a company which lost $327 million on revenue of $775,000 and yet is valued at over $5 billion shows what kind of whackadoodle bubble we are in.
Re: A bubble (Score:2)
Wut? Which one is that? Wowzers.
Cash grab, ride the bubble (Score:3, Insightful)
It's very 1999 dotcom. You just need to hop off the ride before the whole thing comes crashing down to reality. Don't get greedy and ride it to zero.
A few of these AI branded companies will be ok. The rest will be updating their resumes.
Re:Cash grab, ride the bubble (Score:5, Insightful)
Hardly need to go all the way back to 1999. This is the new blockchain circa 2017 [wikipedia.org] one of a number of crypto-related sub-bubbles.
Re: (Score:2)
Blockchain investments circa 2017 might not be the best example at the moment, because it seems like we're in a crypto bubble again in 2024. This seems to be cyclical now... they run up every 3 or 4 years and then come crashing back down.
Re: (Score:2)
There are still many very rich people with politicians in their back pocket who keep blowing air into that particular bubble. And there are still a lot of gullible marks who continue to fall for it. It's actually quite cleverly done. They allow just enough of the marks to make a profit out of it that the rest think they can too. It works a bit like a lottery in this way. 'It could be you!'
This will likely continue for some time yet, until enough politicians grow a backbone and legislate it out of existence.
I do what I did in 99 (Score:5, Insightful)
Me and a few friends made some bespoke POS/CMS/Web Portal applications for the dotcom hopefuls. This was back before off-the-shelf solutions were all over and the few that existed usually needed more configuration than just rolling your own.
We charged insane rates and got them paid, mostly because they believed in their dreams as much as their banks and VCs did, and they wanted their shit on "the internet" as soon as possible, so we worked round the clock. The money was good. Really, really good.
Then the whole shit came crashing down and we went away because we just delivered a product and got paid for it, we had no VCs or banks breathing down our necks and as an added bonus, nobody wanted us to maintain that code.
THIS is how you do it, people!
Re: (Score:2)
THIS is how you do it, people!
Damn right! As an added bonus, you're username matches! Well done all around sir
Re:I do what I did in 99 (Score:5, Insightful)
You don't make a fortune mining gold.
You make a fortune selling supplies to the gold miners.
Re: (Score:2)
Likewise, you don't get rich working, you get rich making others work for you.
Re: (Score:1)
> The money was good. Really, really good.
They promised to pay me partly in stock options instead of a giant paycheck, and we know how that turned out. I couldn't be picky because I had come from the desktop software world and didn't have enough web experience. (Many just lied about web experience on their resume, but I wasn't a good liar.)
Re: (Score:2)
Nope. We only accepted cash-only. And of course they were quite willing, after all they didn't want to share their soon-to-be-billions with us plebs.
Re: (Score:1)
That was perhaps a good bellwether: if they preferred to give one stock options over cash, then they know their biz is dodgy, and the reverse if they believe in their company. Of course egos often lie to owners.
Re: (Score:2)
THIS is how you do it, people!
yeah work won't make you rich, but on the right track you will be ok, and you don't really need any bubble. what you just explained describes my entire career as a self-taught and oldschool "sw engineer": working for clueless idiots with too much money. i even had the luck to work on some groundbreaking and very fun project, although that was not the norm. as an autist i didn't even bother with negotiation. no need, i retired at 50 anyway. now i just code for hobby.
on the plus side i only had to program for
Re: (Score:2)
"You just need to hop off the ride before the whole thing comes crashing down to reality."
Well, that's the trick, isn't it? There's no way to tell exactly when the ride will end until it happens, and then it's too late. It's an intrinsic property of bubbles that the great majority of investors will *not* get out in time. It's a zero-sum game and the odds are decidedly not in your favor.
Re: (Score:2)
It's an intrinsic property of bubbles that the great majority of investors will *not* get out in time. It's a zero-sum game and the odds are decidedly not in your favor.
i learned yesterday that isaac newton, considered one of the brightest minds of humankind, lost a substantial portion of his fortune in a ponzi scheme. fun fact, he did get out and made substantial profit, but then had second thoughts and fomo'ed right back in, losing the full investment. so it seems you can't really be smart enough.
Am I the only one who read it as AI stock ballons? (Score:2)
I'm so disappointed, I thought the story was about balloons with AI, that would have been really cool (but probably useless)
Re: Am I the only one who read it as AI stock ball (Score:2)
Hehehe...
My first read was that "AI is stoking balloons" and thought it had something to do with He shortage...yhea I'm tired today.
dot.com, anyone? (Score:2)
This is gonna take off. Really. You'll just see. We'll make millions. Heck millions, billions, we'll make billions, I tell you!
It's because of the possibilities (Score:3)
Re:It's because of the possibilities (Score:5, Insightful)
The technology has the possibility to replace as much as 20% of the work being done in the next 10 to 20 years. That's trillions of dollars. Even throwing a few hundred billion at that is peanuts.
When AI replaces as much as 20% of human workers, do you think the damage caused by the mass chaos and rioting will be limited to “peanuts”, or will those drunk on “possibilities” finally get it?
Especially when you are ultra wealthy and it's not really your money because of all else fails you know you're holding the entire world hostage and you're too big to fail.
If you’re one of the ultra wealthy more directly responsible for mass unemployment, you’re going to be ultra lucky if you don’t end up with your head on a pike. If a 10% unemployment rate is unthinkable, imagine the chaos of twice that.
Re: (Score:3)
When AI replaces as much as 20% of human workers, do you think the damage caused by the mass chaos and rioting will be limited to “peanuts”, or will those drunk on “possibilities” finally get it?
This. If we lived in the Star Trek post-monetary universe, we could casually talk about replacing 20% of the workforce with robots, because those people would then simply have more free time to work on whatever their passion is - writing, gardening, traveling, researching...
If we lived in The Running Man universe, we could casually talk about replacing 20% of the workforce with robots, because those people are unimportant and can be tossed into pauper camps.
If we lived in The Island univese, we could casual
Re: (Score:2)
You never know. He may revert back to his Democratic New York roots now that he realizes he could actually die instead of going to prison and change his stance to banning all guns. I'd say he wasn't that stupid, but it's pretty hard to tell.
Re: (Score:3)
Oh please. This false flag will have no effect on the convicted felon. He'll use it to show his brainwashed followers he's the second coming as he violates everything that Mexican dude said to do to your fellow neighbor.
The convicted felon has already broken 7 of the 10 commandments, and probably can't even name one of them, and Chris
Re: (Score:3)
But I agree with you that if only big businesses with profit motive control AI development, we are looking at a potentially dystopian future.
Re: (Score:3)
It won't happen overnight, it will be a gradual transition. AI will actually, like most technologies in the past, likely promote economic and productivity growth which is a net job creator.
Gradual? Was 6,000+ layoffs at Twitter ‘gradual’? Is the rate of layoffs in tech now ‘gradual’? Do we often describe the greed-hype and insanity of tech bubbles as ‘gradual’?
Humans are already being laid off at companies so they can invest in the human replacement tech they don’t even have yet. That, is how fast this is happening today. And with every AI “breakthrough” CEOs will become that much more attracted to replacing all those pain-in-the-ass
Re: (Score:2)
Those 6,000 people were laid off because of greed and stupidity, they weren't replaced with AI.
Re: (Score:2)
Those 6,000 people were laid off because of greed and stupidity, they weren't replaced with AI.
6,000 people were laid off, and a business kept running. That’s not greed. That’s Common F. Sense calling out the ignorance and stupidity of former business owners.
What is actually greedy and stupid, is prematurely firing your human workforce for employee replacement technology that isn’t even a viable replacement yet.
Re: It's because of the possibilities (Score:2)
Given the way the labor market is treated like a race to the bottom I would not be so sure about the net job creation. If anything the AI revolution will likely be more similar to the outsourcing craze of the 90s where corporations were looking to cut costs anyway they could by finding the cheapest source of labor. Quality of product does not really matter if you either control the market or collude with companies to keep out new competition.
Re: (Score:2)
Given the way the labor market is treated like a race to the bottom I would not be so sure about the net job creation. If anything the AI revolution will likely be more similar to the outsourcing craze of the 90s where corporations were looking to cut costs anyway they could by finding the cheapest source of labor. Quality of product does not really matter if you either control the market or collude with companies to keep out new competition.
Funny part about Greeds plan here, is knowing Greed is stupid and blind enough to not see the fact that current corporate revenue is only sustained when your human customers have paycheck money to piss away on unnecessary products.
UBI is nothing but Welfare 2.0 with a fancy name. Corporate Greed will eventually realize the BIG bottom line difference between survival income and piss-it-away income when most are filing bankruptcy paperwork because unnecessary products are unnecessary.
Re:It's because of the possibilities (Score:5, Insightful)
People vastly over-estimate the speed at which it will happen though.
I've tried using ChatGPT for work, and it's surprising how it can help, but also very obvious what its limitations are. I fed it a schematic and asked it to check it. I introduced a few trivial errors just as a test. It picked up on 2 out of 3 things I did deliberately, and hallucinated a couple more.
So it's a potentially useful tool for checking things that the normal design rule checks would not detect, but you still have to check the work manually and be able to know what it's confidently talking bollocks.
Re: (Score:3)
I think this is the point. It's not likely to replace many jobs at all. It may slightly improve productivity in some jobs,but not to anything like the extent claimed. And LLMs will never completely replace humans.Or not for long anyway. Companies that try this will continue to get burned when they discover it's hallucination and false positive ratee.
Take the one used to detect cancers from scans, for example. It improves output because it detects cancers earlier than the human eye can, but this comes at the
Re: (Score:2)
The technology has the possibility to replace as much as 20% of the work being done in the next 10 to 20 years.
Before that, it will drive a 40% increase in demand for employees to fix everything the technology fucked up.
I just created a program fix that no automation system could have done. It required a deep understanding of the problem domain, and a deep understanding of where the problem resided. It wasn't at all obvious (and I created the system from top to bottom), and the solution could not be achieved by permutation. It was one of those things I would LOVE to give to an LLM and watch as it generates absolute
When does the bubble go "Pop"? (Score:2)
BubbleGPT (Score:2)
I knew the dot-com and mortgage bubbles would pop soon after people noticed the profit math looked dodgy*, but never figured out a safe and reliable way to profit off those guesses. Stock "puts" and "shorts" are often cited, but have rocket-science fine-print attached to them. Counter-investment products are not designed around consumers.
I have no doubt AI will continue to improve, but perhaps not at the pace or flow many expect. AI has a history of fits and starts.
* I remember one interviewer asked a pro-m
Re: BubbleGPT (Score:2)
Counter investment can be done by investing in stocks that go up when the others go down. Preferably with long term calls. Easy, cheap and no fine print.
Now you just need to understand which stocks go up when a given market crashes.
Suppose AI crashes - people would sell tech. But what would they buy?
An index fund made up of everything-but-tech might be interesting.
If only we could have a fix the climate bubble (Score:3)
Pop! (Score:3)