> Market signals on an impending AI bust are broader than just Oracle’s woes.
It's worse than that - I believe that Oracle is one of the (many) companies right now that, if their AI experimentation fails, will stop the music, and everyone will be running for a chair.
Oracle is one of a few foundational components in the circular-investing group of AI companies. If they fail to make their commitments they're the first domino to fall.
Stay well diversified, keep investing each month, and take a nap.
There are almost surely severe bumps ahead for the AI space and that will likely spill over into the broader market. But unless you’re retiring in the next few years don’t worry about it. You can’t time the ups and downs and the only proven strategy is to just keep investing in a broad indexed portfolio and just ride out. You’ll take a short term hit but also end up buying on the dip because you don’t stop investing.
I suppose I'm just a little worried about a 10 year sideways market. The run-up has been absolutely insane the past year...some graphs are just a literal straight line up. I didn't get to participate in much of that and concerned the prevailing wisdom on these larger timescales may no longer hold true.
> I suppose I'm just a little worried about a 10 year sideways market.
In that case possibly go with something global; in the one recent period where S&P500 was pretty much sideways for 10 years, MSCI World, say, did somewhat better. If the wheels do fall off for the AI bubble, it'll probably hit the US market harder than others.
_Within reason_ (you probably don’t want an index that has literally every stock in the world, say), broader indexes are generally less volatile than narrower ones; must downturns are at least somewhat regional, and sectoral downturns hit some regions harder than others.
The S&P is through the roof because of the AI boom - it's bound to crash if the big players do. What's a better index?
It's hard to imagine a world in which these broad indexes don't crash too. It's the sector indexes I'd love to understand better from a cyclical point of view, so I can buy something that won't also crash.
It's worth adding that conventional wisdom says, you can't time the market. On average, people shifting between cash and stocks to time shocks lose out over just holding a fixed portfolio.
Sometimes conventional wisdom stops being wise. Also 90% of the people in charge of conventional wisdom have their personal wealth depend on retail investors not selling.
I moved 80% of my money out of Vanguard's Target Date Retirement funds and into a money market on June 1st. In the 1.5 months since, the remaining Target Date Retirement fund has fluctuated up and down by about 0.1%. It has basically plateaued. I don't think I am losing out on potential short term gains. I like the idea that I have cash available to buy in on the day of the crash.
Good luck dude! This kind of move can pay off big or not, clearly. I’ve personally talked to fable about this a lot, suggest everyone does.
There are a lot of failure modes. The dot-com bubble looked obvious in 1997; it popped in 2000. Anyone shorting in '97-'98 was carried out on a stretcher before being vindicated. In fact 2000-2002 fell in three brutal legs over two years, and anyone who leveraged up after the first 25% leg was destroyed by the next two.
My boss has already done this several times over the past couple years because of some impeding market crash. Then he goes back and buys a week or so later.
> I moved 80% of my money out of Vanguard's Target Date Retirement funds
which target date fund exactly? You can increase risk/reward buy choosing a target date fund far in the future or you can reduce risk/reward by choosing a target date fund closer to the present. The point of those funds is to gradually reduce your risk as you get closer to your planned retirement date. I moved my 401k into a target date fund about +10 years from my planned retirement (I'm 50). So a little bit on the risk++ side but not much.
What percentage drop do you need to consider it a 'crash' and to buy back in? Is it a drop from when you sold or the peak? What if the peak is another 20% higher and the crash is 10%? While you're sitting in cash, you're also losing to inflation each day.
I generally consider timing the market such as this to be a fools errand, but if you're going to do it you need to have a plan beforehand and follow it. A target date fund does exactly this with allocations.
I don't know, but they aren't really in the same category either. The pandemic didn't shut down everything. It didn't really shut down much, people worked from home and got deliveries instead of doing things in person. There were sectors that were hit bad, but certainly not everything.
The AI crash is about stock market indicator ratios matching those that preceded other major crashes. That's what got me spooked. I don't want to be heavily invested in those companies when/if something bad happens.
My point is that whether there will be a crash or not is incredibly hard to predict. COVID did not come with a stock market crash, but it affected employment much more than a possible AI crash will.
> The AI crash is about stock market indicator ratios matching those that preceded other major crashes.
The way to put faith in such indicators is not (only) by looking at prior crashes, but by forward testing them. Over the last decade, it's been common for me to hear a sentiment like yours: "Indicator X has always resulted in a serious downturn in the past, and we're in X territory now" - and no crash ensued. Over and over again.
Find me an indicator that someone back tested, and then also actually predicted a real crash (with zero false positives). The cost of even a single false positive can be huge. Ask the guys who pulled out (or sold their houses) when COVID struck.
Don't become the person who predicts 7 of the last 2 recessions.
> My point is that whether there will be a crash or not is incredibly hard to predict. COVID did not come with a stock market crash
As someone who had early PUTs against the obvious industries (travel, hospitality) - what I didn't foresee was the insane amounts of government liquidity poured into the markets.
> As someone who had early PUTs against the obvious industries (travel, hospitality) - what I didn't foresee was the insane amounts of government liquidity poured into the markets.
They did it in 2008 as well. Although the amount in COVID seems insanely high, back in 2008, $700B was insanely high. People couldn't believe the government would spend that much to keep the economy going.
Retail investors do not have access to systems that calculate risk, margins, pnl, etc... and generally also don't have the necessary knowledge and market data to price such instruments correctly.
Most ppl are better off KISSing and lowering risk by selling equity for fixed income.
You almost always lose a lot of money if you're seeking safety. Protection from downside risk on your S&P500 investments may cost 20-30% of your investment at which point you're better off just selling the investment and hoping it doesn't go up by that much.
Assuming you are the average person, and not a financial professional, using actual financial hedging instruments properly is unlikely, and far more likely to just increase risk and lower expected return.
A realistic way for an American citizen to reduce risk in the current market is to have a globally diversified portfolio that under-allocates to the US.
> What's the best way to hedge against this, considering many of us have significant savings in the market?
honestly, if you're >= 10 years away from needing that money (retirement or whatever) then the best hedge is to ignore the news and just keep contributing to your investment as always. I got caught up in a couple moments (tarif drama April before last was one) where i panicked and sold and then it only took a few months to get back to even meanwhile 18% of my capital gains were now due to the taxman. I wrote a check to the IRS for 10's of thousands for no reason except over reacting and ignoring every financial advisor's advice.
if you're going to need your investment money within 10 years then you need to get advice on how to start reducing risk (and therefore reward) because you don't have time to survive and repair from a crash.
I knew guys who panicked in Feb 2020, at the start of covid. They moved everything to cash, never got back into the market. Things recovered faster than they thought. The unfortunate truth is they would've more than doubled their money if they stayed invested.
#1: Great question, and I would love to hear the answers (And am learning from the ones posted)
#2: What I've done so far: Haven't bought stock in a year. Have moderate short positions on Palantir, SpaceX, and Tesla. Have big short positions in the most popular Quantum computing companies. (Scams IMO). I have sold most of my positions ("profit taking"?) in stocks which have gone up a lot in the past year. (Nvidia, Broadcom etc), and am no longer using margin; about 1/3 of my brokerage value is now "cash", generating ~3% interest.
Wouldn't it be wiser to get out of the market into fixed rate assets like government bonds? Maybe have some into puts on SPY (or QQQ since tech would probably have bigger losses) too, but mainly getting out of long positions on what seems a really overvalued stock market
I just hope the companies aren't considered too big to fail. Bailing them out would be a bad idea.
They will be. When the SHTF, you'll see Rubio in the room^H^H^H^H circus tent, sitting right next to Bessent, arguing that propping up OpenAI is as much a national security interest as bailing out GM was.
Just sell all your ETFs and buy them again when the market goes up or down. You're very likely to lose money with options and you will definitely lose a lot of money if you buy enough options to hedge your full exposure.
You will definitely lose less in opportunity cost than the actual cost of hedging your position, because hedging is extremely expensive and cancels out almost all gains. If it was cheap, everyone would do it.
unless you're doing this in an IRA or your 401k remember the IRS wants its cut of any gains you may lock in. That's a painful check to write let me tell you.
Reminder: Serious people have been predicting a market crash "within the next 3 months" for 3 years now. In that time, the "market" has gone up around 70% (66%-86% depending on the what part you are looking at).
A friend of mine and I go out to lunch every 3 months and talk about, among other things, investing. We've made a trope of it, calling out the people who are predicting an imminent market crash every time we have lunch.
I'm not saying that it doesn't look like it's going to crash, but I'll also say that there's also a very sizeable downside potential for getting out of the market.
It's tempting to sell a bunch, but then you've got cash. What do you do with cash when the government keeps printing money and assets are all overpriced?
You think the hedge funds selling SPY options don't have this priced in already? Of course, you can still make money on this bet, just like you can win money at a roulette table, but unless you think have some special insight that hedge/quant funds don't have, buying options should be negative EV.
Options market makers have no idea where the S&P will be in one year, options are priced on the current implied volatility. The bid and ask will be slightly lower and higher than the true current option price so the MM can make their nut on the spread and then hedge so they’re delta neutral.
If you buy a put you are making a bet that realized volatility will exceed implied volatility. This may or may not happen and there’s no way to predict the future.
My understanding of the ai circular financing racket is that not everyone will be running for a chair.
Nvidia owns all the chairs, and they’re letting other companies pretend to for a while, but if it all falls apart the backstop to the collapse will be nvidia.
Doesn't Nvidia's success depend hugely on AI money pumping up demand for their products? If/when AI companies run out of money to keep investing in data centres, the bottom will fall out of the market and hopefully we can go back to buying reasonably priced graphics cards.
Takes a lot of IaaS to support the GPUs and workflows, all of that kit is immediately re-useable as general purpose compute to exit the commercial DCs they operate out of today.
Much of their current debt fuelled expansion isn't singular to AI. The circular narrative ignores this.
I think he's saying that if one company panics or fails, all of the other AI companies will quickly pull out, and it will be a race to not be holding the hot potato at the end when the bubble bursts. Lots of analogies there! This is where we get the phrase "the stock market takes the stairs up and the elevator down". It crashes quickly and hard, and can be triggered by one event.
The qemu-img(1) installed on my system claims that it supports every disk format supported by Virtualbox [0], so I guess the only thing left would be to be able to handle the "machine definition" file.
qemu definitely won't do that out of the box, so, yeah, VirtualBox is better than qemu there. But I bet there's a fancy-pants GUI out there that has an import wizard that will handle that for you.
Regardless of how suitable qemu is for Vbox users, it's unmatched for batch uses, like as something part of a distro build chain to allow non-cross-compilable packages to be built for ARM on x86-64 and such. And ... to interactively step into cross-compiled sysroots.
Vbox has the desktop experience. The "guest integration" stuff and whatnot. If you've used it for years, familiarity.
Quite possibly. I know that it's what Openstack uses for virtualization.
> ...to interactively step into...
Sure, running on an emulated CPU can be very nice.
> ...like as something part of a distro build chain to allow non-cross-compilable packages to be built for ARM on x86-64...
This sounds like it would be incredibly slow. As in "It's better to just buy some inexpensive ARM PCs" slow. I know that GCC and Clang can cross-compile just fine. It looks like Rust cross-compiles, and I think Go does as well. Do you know of major tools and/or languages that don't cross-compile?
Oracle has been a toxic tech for a long time (along with IBM). I don't think anyone is putting it in the same bucket of modern tech companies, let alone AI companies.
Frankly, their forays into dubious financial engineering and investments are expected at this point.
Everyone in the tech and media world is dead set on this being a bubble.
Yet, even now, Fable is able to do the work of 4-5 engineers when used by a single senior engineer. Teams can and will shrink.
Look at all the production and advertising companies switching over to Seedance. I know ad firms bidding 1/4th their typical contract price (pharma, P&G, etc.) and winning contract after contract.
This isn't dotcom "dark fiber" before demand. The demand is here now, big legacy firms are just struggling with deploying it. Nimble small teams are making a killing.
A financial bubble has almost nothing to do with how good the product is. It's about how much of the value the company can capture, and what the ratio of that capture is compared to the investment.
It doesn't matter to investors if OpenAI or Anthropic can build AGI if a year later 10 competitors have similar models and eat into the revenue. OpenAI and Anthropic needs years, if not decades, of significant market dominance, post-enshitification, to justify their investment spend.
> Yet, even now, Fable is able to do the work of 4-5 engineers when used by a single senior engineer. Teams can and will shrink.
If that's true or not, it's a bit irrelevant. Maybe teams won't shrink because of Jevon's paradox, or maybe tech debt will catch up.
But it doesn't matter because the people calling this a bubble mostly believe that the companies burning money cannot have the return on investment needed. This can be for a variety of reasons, but my favourite one is just that open source AI models are good enough, cost a fraction of what the frontier ones do (with predictable costs), can be fine tuned, and can be relied upon (no orange tweet banning your acces to the model you've been using). So for me OpenAI and Anthropic will really struggle to merit their valuations.
And then companies like Oracle are just a dumpster on fire. GPU hosting is a commodity business; expensive one, for sure, but there's no way in hell they'll make actual returns on the money burned with zero moat. And things are even worse when you consider the political involvement of the CEO and his nepo baby, which can easily burn good will.
> Everyone in the tech and media world is dead set on this being a bubble.
is completely orthogonal to this:
> Yet, even now, Fable is able to do the work of 4-5 engineers when used by a single senior engineer.
The industry being in a bubble or not is irrelevant to the tech being good or bad. The dot-com bubble popped (and was a bubble) even while the tech was fit for purpose.
I think the "bubble" is more about return on investment and not usefulness of the technology. So much money has been invested on the assumption that so much return is going to materialize. The more money going in the bigger the expectation of return, that's the bubble.
Yes, but all bubbles (except the tulips...) have a real, valuable, new technology at their core. That it's amazing technology doesn't stop the financial side of it being a bubble. In fact it all but ensures it is.
It's worse than that - I believe that Oracle is one of the (many) companies right now that, if their AI experimentation fails, will stop the music, and everyone will be running for a chair.
Oracle is one of a few foundational components in the circular-investing group of AI companies. If they fail to make their commitments they're the first domino to fall.