I'd be willing to bet that, like many US websites, OpenAI's users are at lest 60% American. Just because there's 20x more people out there doesn't mean they have the same exposure to American products.
For instance, China is an obvious one. So that's 35%+ of the population already mostly out of consideration.
>It’s also obvious that they’re selling heavily to businesses, not consumers.
I don't think a few thousand companies can outspend 200m users paying $200 a month. I won't call it a "mathematical impossibility", but the math also isn't math-ing here.
Even if you grant that OpenAI might be as successful as Apple at international expansion and support, that’s still only a non-US market about double the size of the US market.
>I know this is the latest catastrophizion meme for AI companies, but what is it even supposed to mean?
Someone else put it succintly.
"When A million dollar company fails, it's their problem. When a billion dollar company fails, it's our problem"
In essence, there's so much investment in AI that it's a significant part of the US GDP. If AI falters, that is something that the entire stock market will feel, and by effect, all Americans. No matter how detached from tech they are. In other words, the potential for the another great depression.
In that regard, the government wants to avoid that. So they will at least give a small bailout to lessen the crash. But more likely (as seen with the Great Financial Crisis), they will likely supply billions upon billions to prop up companies that by all business logic deserved to fail. Because the alternative would be too politically damaging to tolerate.
----
That's the theory. These all aren't certain and there are arguments to suggest that a crash in AI wouldn't be as bad as any of the aforementioned crashes. But that's what people mean by "become too big to fail and get bailed out".
The closest analogy is the dot-com crash and there really wasn't any bailout for that, despite the short term GDP impact. And billion-dollar companies were involved back in the day too, like Apple, Microsoft, Amazon, Ebay etc. etc.
If they aren't dumb, why are they investing in MSFT now then if it's a bubble that's doomed to fail? And even in the worst case scenario, a 10-15% decline in the S&P 500 won't trigger the next Great Depression. (Keep in mind that we already had a ~20% drawdown in public equities during the interest rate hikes of 2022/2023 and the economy remained pretty robust throughout.)
Like I said, they aren't "that" dumb. They are playing a risky game, but when they see the number go down rapidly they will pull. Which will make the line go down even faster.
>And even in the worst case scenario, a 10-15% decline in the S&P 500 won't trigger the next Great Depression
Only if you believe the 10% decline won't domino and that the S&P500 is secluded from the rest of the global economy. I wish I shared your optimism.
> and the economy remained pretty robust throughout.
Yeah and we voted the person who orchestrated that out. We don't have the money to pump trillions back in a 2nd time in such a short time. Something's gonna give, and soon.
> Only if you believe the 10% decline won't domino and that the S&P500 is secluded from the rest of the global economy. I wish I shared your optimism.
So your hypothesis is that a 10% decline in the S&P 500 will trigger the next Great Depression, i.e. years of negative GDP growth and unemployment? I agree that it could cause a slight economic slowdown, but I don't think AI and tech stocks are a large enough part of the economy to cause a Great Depression-style catastrophe.
The problem is that the non AI economy is already in the toilet. The consumer and commodities markets are all flashing red. Consumer debt is all time high. Inflation is still punishing the bottom end of workers severely and the ACA cuts will cause a lot of financial stress (unless people of course discountinue their plans)
An expected outcome from a AI blowout is the uncertainty and everyone holding onto their assets and credit recalls plus interest rate hikes.
During the great depression it wasn't the stock market collapse that caused it as much as it was the credit crunch that followed. Prior to the blowout people literally bought stocks on credit.
>So your hypothesis is that a 10% decline in the S&P 500 will trigger the next Great Depression, i.e. years of negative GDP growth and unemployment?
Yup. I won't say it's the only factor, nor biggest. But I'm focusing on this topic and not 40+ years of government economic abandonment of the working class. It's the straw that will break the camel's back.
What "we" know already is hard to add to, as a forum that has a dozen AI articles a day on every little morsel of news.
>whereas OpenAI has a clear opportunity with advertising.
Personally, having "a clear opportunity with advertising" feels like a last ditch effort for a company that promised the moon in solving all the hard problems in the world.
Going viral is great when you're a small team or even a million dollar company. That can make or break your business.
Going viral as a billion dollar company spending upward of 1T is still not sustainable. You can't pay off a trillion dollars on "engagement". The entire advertising industry is "only" worth 1T as is: https://www.investors.com/news/advertising-industry-to-hit-1...
because these are mostly the same players of the 2010's. So when they can't get more investor money and the hard problems are still being cracked, the easiest fallback is the same social media slop they used to become successful 10-15 years prior. Falling back on old ways to maximize engagement and grind out (eventually) ad revenue.
It did. I question the issue of "what problem am I trying to solve" with AI, though. Transportation across a huge swath of land had a clear problem space, and trains offered a very clear solution; created dedicated railing and you can transport 100x the resources at 10x the speed of a horseman (and I'm probably underselling these gains). In times where trekking across a continent took months, the efficiencies in communication and supply lines are immediately clear.
AI feels like a solution looking for a problem. Especially with 90% of consumer facing products. Were people asking for better chatbots, or to quickly deepfake some video scene? I think the bubble popping will re-reveal some incredible backend tools in tech, medical, and (eventually) robotics. But I don't think this is otherwise solving the problems they marketed on.
This is a use case that hasn't yet been proven out, though. "Good enough" for an executive may not be "good enough" to keep the company solvent, and there's no shortage of private equity morons who have no understanding of their own assets.
>That doesn't mean AI is going to go away, or that it won't change the world - railroads are still here and they did change the world - but from a venture investment perspective, get ready for a massive downturn.
I don't know why people always imply that "the bubble will burst" means that "literally all Ai will die out and nothing will remain that is of use". The Dotcom bubble didn't kill the internet. But it was a bubble and it burst nonetheless, with ramifications that spanned decades.
All it really means when you believe a bubble will pop is "this asset is over-valued and it will soon, rapidly deflate in value to something more sustainable" . And that's a good thing long term, despite the rampant destruction such a crash will cause for the next few years.
But some people do believe that AI is all hype and it will all go away. It’s hard to find two people who actually mean the same thing when they talk about a “bubble” right now.
I don't think anyone seriously believes AI will disappear without a trace. At the very least, LLMs will remain as the state of the art in high-level language processing (editing, translation, chat interfaces, etc.)
The real problem is the massive over-promises of transforming every industry, replacing most human labor, and eventually reaching super-intelligence based on current models.
I hope we can agree that these are all wholly unattainable, even from a purely technological perspective. However, we are investing as if there were no tomorrow without these outcomes, building massive data-centers filled with "GPUs" that, contrary to investor copium, will quickly become obsolete and are increasingly useless for general-purpose datacenter applications (Blackwell Ultra has NO FP64 hardware, for crying out loud...).
We can agree that the bubble deflating, one way or another, is the best outcome long term. That said, the longer we fuel these delusions, the worse the fallout will be when it does. And what I fear is that one day, a bubble (perhaps this one, perhaps another) will grow so large that it wipes out globalized free-market trade as we know it.
Bubbles bursting aren't bad unless you were overinvested in the bubble. Consider that you'll be wiping your ass with DIMMs once this one bursts; I can always put more memory to good use.
> Bubbles bursting aren't bad unless you were overinvested in the bubble.
That's what I am trying to say: every big technology player, every industry, every government is all in on AI. That means you and I are along for the ride, whether we like it or not.
> Consider that you'll be wiping your ass with DIMMs once this one bursts; I can always put more memory to good use.
Except you can't, because DRAM makers have almost entirely pivoted from making (G)DDR chips to making HBM instead. HBM must be co-integrated at the interposer level and 3D-stacked, resulting in terrible yield. This makes it extremely pricy and impossible to package separately (no DIMMs).
So when I say the world is all in on this, I mean it. With every passing minute, there is less and less we can salvage once this is over; for consumer DRAM, it's already too late.
Games tend to avoid FP64 compute as Nvidia has always gimped it in consumer GPUs, so you are somewhat lucky there. "Lucky" as in, you get to enjoy the broken-ass, glitchy FP32 physics that we've all grown to love so much.
However, if you actually need the much higher precision of FP64 for scientific computing (like most non-AI data center users do) and extremely slow emulation is not an option, consider yourself fucked.
My only critique is that it would help to group projects by difficulty. But AI genned or not, it does have decent ideas and the follow ups I clicked on for "getting started" all at least seem non-AI genned. As some examples of what I have done myself previously, Linking to Shirley's "Ray Tracing in a Weekend" for a Ray tracer seems pretty solid, but throwing the GBATek manual at you for making an emulator is very "the rest of the owl" sorts of advice.
If it at least inspires some people to actually get their hands dirty (instead of outsourcing their intelligence to a black box), I don't mind Ai being used as a brainstorming tool.
It is not that i am against AI genned articles, it is that this is just a very low-effort one. They could have easily categorized, noted difficulty levels, background needed, added more details/pointers etc. to really make this useful.
For example, to mention implementing TCP/IP stack but not point to Adam Dunkels works is a unforgivable crime ;-) See list of his projects at https://en.wikipedia.org/wiki/Adam_Dunkels#Networked_embedde... which are all relevant here. He is another Fabrice Bellard like super programmer!
I think people should try to implement one or more of these projects with AI tools but using it solely for drudgery work i.e. keep design/architecture in your head (this is where the AOSA books come in) and only get AI to generate code for well specified (formally if possible) functions. This will also make them highly relevant to today's job market. Programmers now need to educate themselves and move more into meta-level i.e. use Formal Methods (Specification/Verification) to get AI to generate code with correctness proof along with it.
That's everything at some point, no? Easier to talk about something, harder to start something, and much harder to actually see it through.
Personally, I am making a ray tracer project right now in Rust, so I hope I can become the latter. It being something I did before (albeit, long ago in c++) helps.
I'd be willing to bet that, like many US websites, OpenAI's users are at lest 60% American. Just because there's 20x more people out there doesn't mean they have the same exposure to American products.
For instance, China is an obvious one. So that's 35%+ of the population already mostly out of consideration.
>It’s also obvious that they’re selling heavily to businesses, not consumers.
I don't think a few thousand companies can outspend 200m users paying $200 a month. I won't call it a "mathematical impossibility", but the math also isn't math-ing here.
reply