It’s always a crowd pleaser to be skeptical of ai development. Not sure what people feel like they are achieving for continually announcing they aren’t buying it when someone claims they’ve made effective use of these tools.
Nvidia is giving OpenAi money (through investment) to buy Nvidia chips. The bubble is that Nvidia got that money from its crazy high stock price, the extra investment raises OpenAi’s evaluation and the increased sells raises Nvidia’s evaluation. If the valuations see a correction then spending like this will decrease, further decreasing valuations.
Bubble collapsing looks like enshittification of OpenAI tools as they try to raise revenues. It’ll ripple all throughout tech as everyone is tied into LLMs, and capital will be harder to come by.
> The bubble is that Nvidia got that money from its crazy high stock price,
This is totally False, NVDA has not done any stock offerings. The money is coming from the ungodly amount of GPUs they are selling. In fact they are doing the opposite, they are buying back their stock because they have more money that they know what to do with.
A company buys back its stock if it thinks the stock is underpriced. Otherwise when “you have more money than you know what to with” you give it to your shareholders via a dividend. A concept mostly forgotten by tech companies.
NVDA outstanding shares are down ~1.2% year over year; the company has been buying back its own shares with —>> profits <<— to the tune of tens of billions.
Meanwhile NVDA stock is mildly up on this news, so the current owners of NVDA seem to like this investment. Or at least not hate it.
Agreed that we’ll see ad-enabled ChatGPT in about five minutes. What’s not clear is how easily we’ll be able to identify the ads.
Valuations won’t see a correction for the core players, I have no idea why people think that. Both of these companies are already money factories.
Then consider we are about to lower interest rates and kick off the growth cycle again. The only way these valuations are going is way up for the foreseeable future
And it's worth reiterating that a bubble does not mean the technology is worthless. The dot com bubble collapsed despite the internet being a revolutionary technology that has shaped every decade since. Similarly LLMs are a great and revolutionary technology, but expectations, perception and valuations have grown much faster than what the technology can justify
These hype cycles aren't even bad per se. There is lots of capital to test out lots of useful ideas. But only a fraction of those will turn out to be both useful and currently viable, and the readjustment will be painful
I think ultimately the conclusion that we're in a bubble is bad analysis. It jumps over a chasm and assumes that analogy to past historical situations allows us to draw conclusions.
This isn't a bubble. This is the collapse of 300 years of modern capitalism into corporate techno feudalism.
This won't crash and lead to a recession or depression. We are at the end game. Look around you. Capital is going scorched earth on labor. They are winning. Cost of living in metropolitan areas is exploding, and most of us will end up begging for scraps in peripheral areas.
This is the result of everything the elites have been working towards for the past few decades. Climate catastrophe is the cherry on the cake: they will shock therapy us into the last few bits. There will be corporate citizenship that enables one to live as a demi-god at the behest of the owners, and survival in the wastelands for the rest of us.
I think everyone is underestimating the advancements in wafer tech and server compute over the last decade. Easy to miss when it’s out of sight out of mind but this isn’t going anywhere but up.
The current SOTA is going to pale in comparison to what we have 10 years from now.
> I think everyone is underestimating the advancements in wafer tech and server compute over the last decade.
What advancements?
We have done a fabulous job at lowering power consumption while exponentially increasing density of cores and to a lesser extent transistors.
Delivering power to data centers was becoming a problem 20 ish years ago. Today Power density and heat generation are off the charts. Most data center owners are lowering per rack system density to deal with the "problem".
There are literal projects pushing not only water cooling but refrigerant in the rack systems, in an attempt to get cooling to keep up with everything else.
The dot com boom and then Web 2.0 were fueled by Mores law, by Clock doubling and then the initial wave of core density. We have run out of all of those tricks. The new steps that were putting out have increased core densities but not lowered costs (because yields have been abysmal). Look at Nvidia's latests cores, They simply are not that much better in terms of real performance when compared to previous generations. If the 60 series shows the same slack gains then hardware isnt going to come along to bail out AI --- that continues to demand MORE compute cycles (tokens on thinking anyone) rather than less with each generation.
child care policy frees labor capacity for work that is more likely to earn a slice of the national income. It’s almost certainly going to result in greater economic activity for the state. In the immediate it is funded from two existing funds.
State + local tax burden in NM is 10.2%[1]. Revenue neutral would mean those taking the child care would instead take a job with average salary $120,000. But as another comment points out this policy attracts new jobs to the state, which complicates the math
Given how much capital has already been committed to infrastructure development I find it very unlikely that even a Black Monday style market correction will do much to alter the medium and long term outlook for ai development and investment.
Too many people talk about a bubble almost wish-casting that such a thing will make this all go away. It’s probably safest to assume your political enemies won’t be hoisted on their own petard anytime soon.
> Given how much capital has already been committed to infrastructure development I find it very unlikely that even a Black Monday style market correction will do much to alter the medium and long term outlook for ai development and investment.
So let's bet on sunken cost fallacy? That didn't work out during the dotcom era.
When people talk about the bubble is mostly about ending the LLM hype, not making everything associated just disappear. The infrastructure, ideas and developments will stay and probably end up being a lot more useful without the noisy hype surrounding it right now.
Yep, but the dotcom burst was still was a burst. Ex post success does not matter in that analysis, and in fact, I'm arguing that the dotcom burst was instrumental for companies like Facebook, Youtube and Google to develop as they did.
A bursting bubble doesn't instakill an entire industry. No one is suggesting that. Bursting bubbles are harmful to most market participants and often have a blast radius beyond the industry. It's great that Facebook and YouTube emerged (I suppose) but that doesn't help grandad who lost 35% of his retirement fund.
> Given how much capital has already been committed to infrastructure development
Here’s the stupid thing about that infrastructure: This isn’t railroads. This isn’t skyscrapers. This isn’t underseas fiber optic. These are, at the end of the day, data center GPUs that can and will start failing en-masse on normal hardware timelines.
When that happens, they will all need to be replaced, at whatever rate NVIDIAs selling at - or can sell at, if there’s any geopolitical incidents. If there aren’t enough customers, capacity shrinks, and the remaining customers must foot the bill for the new system and the previous round of investment, causing a death loop wherever there isn’t profitable use feeding into the system.
A clearer comparison: It’s like if airlines had to replace their entire fleet every 3-4 years while selling $20 tickets to anywhere; and the growth metrics look incredible when everyone’s flying at that rate. So much so, that investors paid for the first fleet round calling it “infrastructure.” Meanwhile the amount of ticket sales keeps growing when everyone finds amazing use cases and new business models from the $20 tickets. Surely, this can only get better from here, and justifies even more fleet investment.
The problem is it will create a broader economic problem in the markets.
The top 10 holdings for VTI, the "easy button" investment option. Are: NVIDIA, Microsoft, Apple, Amazon, Facebook, Broadcom, Google, Tesla, Berkshire Hathaway. They represent a third of US market value, and 60% of them are heavily hooked into AI. A panic will be a decimation for NVIDIA, and that splash zone will be broad.
Totally! You can also use something like VIG or VDADX if you're restricted in terms of offerings.
I personally account for that in my planning. But many folks have their money in 401k funds in generic broad market securities -- often with high fees on top of it!
Ok so they sort of have something in common...but I'm not sure it's good advice to say one can be used to replace the other due to their very different overall strategies.
I mean, every year or so there's a new NVIDIA GPU, making last years models obsolete and GPUs from 2 generations back essentially landfill. I don't think there's another large-scale investment that decays at this rate.
The infrastructure being there might also be a problem.
- If there's no breakthrough, and we're only going to make incremental gains, then its wasteful
- If there's a breakthrough, it might turn out a new AI architecture needs either lot less compute, or a different kind of it
> I mean, every year or so there's a new NVIDIA GPU, making last years models obsolete and GPUs from 2 generations back essentially landfill. I don't think there's another large-scale investment that decays at this rate.
Ampere generation A6000’s are still selling used for close to MSRP on eBay. I bought an A4500, used it for a year and then sold it for MORE than I bought it for.
I’ve never seen computer hardware appreciate in value like it is now, even if new cards are still significant upgrades.
Yeah, I think it's pretty messy. I bought some keplers a year ago (far below MSRP), and they are fine -- but the trouble becomes software support; sort of like tablets/phones where the hardware's still more than enough, but because software stops supporting it, the landfill becomes landfilled -- excepting those people willing to set up the very specific environments where there's still use in them -- and to that end, obviously the things're more than a decade old, but hardly useless, though it would be fair to say most people would not run them.
If you can keep up software support, the hardware becomes much more valuable both now and later, but this is much harder to do in the consumer space where the software than in the industrial/AI space; I'd say the majority of my PC replacements until I was established financially were the result of software not supporting my hardware (not that it couldn't do the task if there were support, but simply that there was no support), and I should say, too, that in almost every case, I regretted dumping $1k in today-dollars for a single game or application otherwise.
Obviously, the problem's much worse with phones/tablets where most people use them for Cloud activities; people use them for 2-3 years and they go in the trash or their service provider might buy it back essentially as e-waste to be scrapped (despite marketing implications of them being reused to help surgeons in Africa).
Clearly, a lot of people very desperately want AI tech to fail and disappear. And if there is such a strong demand for "tell me that AI tech will fail", then there will be hacks willing to supply.
Now hold my beer, as I cast a superfluous rank to this trivial 2nd order Tensor, because it looks awesome wasting enough energy to power 5000 homes. lol =3
This isn't exactly the case. The trend is a log scale. So a 10x in pretraining should yield a 10% increase in performance. That's not proving to be false per say but rather they are encountering practical limitations around 10x'ing data volume and 10x'ing available compute.
> (Or at least because O(log) increases in model performance became unreasonably costly?)
But, yes, I left implicit in my comment that the trend might be “fleeting” because of its impracticality. RL is only a trend so long as it is fashionable, and only fashionable (i.e., practical) so long as OpenAI is fed an exponential amount of VC money to ensure linear improvements under O(log) conditions.
OpenAI is selling to VCs the idea that some hitherto unspecified amount of linear model improvement will kick off productivity gains greater than their exponentially increasing investment. These productivity gains would be no less than a sizeable percentage of American GDP, which Altman has publicly set as his target. But as the capital required increases exponentially, the gap between linearly increasing model capability (i.e., its productivity) and the breakeven ROI target widens. The bigger model would need to deliver a non-linear increase in productivity to justify the exponential price tag.
