This is a LONG article, covering a lot of ground, so the small snippets below only suggest at the details and rabbit-holes in the full piece.
If you're interested in AI, the economy, how the stock market works (and WHO it's working for), or . . . simply enjoy honest journalism, this article is a gold mine.
Oh, and speaking of OpenAI, check out what Apple's doing in response to that company's bad behavior.
And it all starts with a question: how much do you trust Sam Altman? The stock market and (to some extent) the global economy rests on your answer.
You see, OpenAI has become one of the largest liabilities in recent economic history. You can argue that OpenAI’s no longer the focal point of the AI bubble — you can talk all you want about open source models or Anthropic or any number of other elements — but without OpenAI, the AI industry doesn’t exist, and the justification for trillions of dollars of capex evaporates.
The AI bubble isn’t a result of any actual return on investment — whether that be in purely monetary terms, like revenue or profitability, productivity gains, or anything tangible or measurable. Rather, it’s an episode of cult-like psychosis that infected the brains of some of the most powerful and wealthy individuals and institutions, where the powerful mythology of a company inspired — and been used to inspire — the greatest capital misallocation in history.
As much as this’ll piss some people off, I fully believe that the only reason this has kept going so long is that OpenAI has yet to collapse. Its failure would be a watershed moment — the Lehman Brothers of the AI bubble, and an event that would define the end of one epoch, the start of another, and that would shake the afflicted out of that psychosis. Absent this wake-up call, NVIDIA has continued to sell GPUs, the coffers of the semiconductor industry have continued to swell, and more and more spending commitments have been made.
. . . The AI trade is the natural endpoint of an increasingly-enshittified stock market where many analysts and journalists exist only to repeat narratives to influence stock prices. Outside of semiconductors, the AI trade has never, ever been about the actual underlying economics or the actual economic potential of Large Language Models, but projecting shadows on the wall to resemble something that looks like the next generation of technology.
. . . That’s because the AI trade exists only to extract value and con investors. It is not a trade related to the actual fundamentals of whether AI works or not, whether AI actually makes anyone money, or really anything about AI at all outside of whether mentioning AI or an AI-related company makes a stock number go up or down.
I’ll be blunt: modern journalism has failed the retail investor and directly helped the wallet inspector regulate the stock market. By empowering Sam Altman and the rest of the AI industry’s deliberate attempts to obfuscate the actual economics of generative AI and setting the terms of AI’s success as “how stocks are doing and whether the companies are growing in general,” they have defaulted on their responsibility to the general public and helped the already-rich get richer.
None of this would be possible if business journalism actually saw themselves as having a responsibility to give their audience good information.
(More. Lots more . . .)
IMHO, the first question everyone should start with is, if we are looking forward to the Great American Prosperity, "what Technological progress has the best potential to completely change our world in the next 20 years?"
Second question should be: "How do we take control of this technology so that it works for us and not for the enemy?"
For sure it's a bubble. Grab the highly subsidized tokens while you can, and build something that can make you money.
Reading through it, I would say, it is an interesting start-point.
In essence, the argument here is tulip-mania.
What if, it is not. What if it is part of a wider development? When put into a wider context:
I am not sold on the premise: tulip-mania.
Two things can be true at once.
The market aspect is undeniable.
I believe you are correct that military and other aspects, mostly behind the curtain, are true also.
Does that mean there won't be negative market consequences?
There already ARE, of course -- anyone wanting or needing to buy electronics today -- phones, computers, gaming consoles, SSDs, ANYTHING with memory -- is paying significantly higher prices than they would have last year, because there just ISN'T enough available memory at any price. The memory famine is undeniable.
We can only guess at the future, but despite the usefulness of AI to Space Force and a hundred other US gov't groups, I suspect we're looking at serious market disruption for a while, and that many US investors (including just people with retirement plans of almost any sort) will be impacted.
True. In terms of consumption:
Just use the phones already available. You would not need the latest gadget.
Use an older phone, root it, stick a new rom on it, and escape closed source BS and all the telemetry that is built into it. Choose apps with care.
Being mindful of what to use is much better. Need a phone? Buy a flip phone. Use a laptop / desktop to do other things. Refuse to be hooked up on systems that require a smartphone based access process.
For instance, you can have keepass to do topt/2FA. You do not need google authenticator.
Banks want you to use an app. Use the calculator, or preferably, move into bank accounts based on stable coins. Lay down your own rails. Customer card in apps: try and figure out if the customer card works on a qr code or an nfc chip. If the latter, use a token programmed for that reason. If the former, then print the QR code, plastify it, and shop away.
Throw out the "smart" from the smart TV. All you need is a laptop and a secondary screen. There are enough media centers running on desktop/ laptop.
You need storage? Hunt for old server disks, and combine them on an USB hub.You do not need networked stuff. To watch a movie or surf the internet, you do not need DDR5. You could make do with DDR3 machines, unless of course, the sights are on 4K and higher screen resolutions and you need to re-compile large files. Although the latter is a matter of time allocation, and there is no reason to waste it waiting for the recompile to finish.
Do I as a consumer really care whether or not my laptop boots into graphical in 30 seconds or 120 seconds just because I am equipped with a NVME card instead of an ssd? Why would I be less satisfied with a 500mb/sec transfer than 1G/s transfer?
Again, it is allocation of time.
It is important to understand that the whole AI growth is resting upon customer profiling. Break it!
Think about deploying Raspberry Pi's for server applications instead of locking yourself into the laptop/desktop paradigm.
Look for p2p: i2p, ifps solutions.
So, I think, the consumer impact is overrated, and is not geared towards existing solutions that work but have not yet fully been deployed. Voting with your data and your wallet ..... is very powerful.
What it might reveal is the bloat that has been stacking up in consumerland vendors. Mal-investment.
And that is a good thing. Demand and supply are influencing prices. Price too high. Allow the demand to plummet. It is an aggregate choice.