Bryan McMahon wrote about a potential AI bubble in The American Prospect.
In 2024, [Apple, Nvidia, Tesla, Alphabet, Meta, Amazon, and Microsoft] were responsible for the lion’s share of the growth of the S&P 500, with the returns of the other 493 companies flat.
For the tech industry, DeepSeek is a threat to its incredible bull run because it proved three things.
First, frontier AI models could be trained much more cheaply and efficiently than the current Silicon Valley approach of building massive models requiring hundreds of thousands of GPUs to train. From a capital perspective, the U.S. strategy is wasteful, relying on at least ten times the investment to make similar model progress.
Second, DeepSeek showed you could train a state-of-the-art model without the latest GPUs, calling into question the current demand for the latest GPUs that is so hot customers have been facing delays of six months to a year to get their hands on them.
Finally, the high valuations of leading AI startups depend on a technical lead in their models to charge prices anywhere near what they need to recoup their computing costs, but that technical lead, enabled by a combination of closed-source models, billions in capital expenditure, and export controls blocking Chinese companies like DeepSeek from accessing the latest GPUs, is gone. Should demand for GPUs fall or even not hit the exponential increases the billions invested are betting on, the bubble will pop.
Between VCs, Big Tech, and power utilities, the bill for generative AI comes out to close to $2 trillion in spending over the next five years alone. […] While AI-fueled coding could definitely boost productivity, it’s hard to see how it could become a multitrillion-dollar industry.
I’ve been reading the (sometimes) bi-weekly newsletters at wheresyoured.at for some time. It’s a very interesting dig into the financials of the companies riding the generative AI wave.
I don’t agree with everything Ed says, but there is definitely something very wrong here. I don’t doubt the market is going to crash when the AI bubble bursts. I doubt LLMs will disappear for good.
There’s a lot of good stuff, but this one really stuck with me:
Leading the round is SoftBank, which is also committing $19 billion, as well as creating a joint venture called SB OpenAI Japan to offer OpenAI’s services to the Japanese market, as well as “spending $3 billion annually to use OpenAI’s technology across its group businesses” according to the Wall Street Journal.
On top of that, OpenAI anticipates it will burn as much as $40 billion a year by 2028, and projects to only turn a profit “by the end of the decade after the buildout of Stargate,” which, I add, is almost entirely dependent on SoftBank, which has to take on debt to fund both OpenAI and the project required to (allegedly, theoretically) make OpenAI profitable.
Freedomcan mean the freedom for everyone to live healthy, productive, and meaningful lives; or it can mean the freedom for one powerful person to make decisions that harm everyone else.
There always has been more demand for tech workers than supply. If a programmer does not like their job then they can quit and find a better one the same day.
Meaningful work and the power to push back against bad leadership have been significant reasons why coders stay at their current jobs.
So, the reason why executives are excited and willing to spend trillions of dollars is so they can fire those skilled workers and replace them with poorly educated workers that supervise machine learning models: powerless workers who can be fired and replaced at will.
Then the executives can make decisions without anything resembling democracy.
Now, in my own voice, if that was your point of view and if your ego was that big then there is no amount of money you can spend that is too much because: b) this is what you always wanted, and a) you don’t want to be on the wrong side of that chasm of inequality this opens up.
hmmm… seems like that open-source model might make some sense after all - who knew?
frankly, i’d be happy to see the U.S. model collapse, again, since it wouldn’t be the first time “AI” ended up in the dumpster after the hype train ran out of rail
meanwhile, the energy requirements cannot be met, and though i haven’t been keeping up with AI rollout, the solution, far as i know, is to build a bunch of nuclear reactors … another bright idea from the minds of bi-pedal primates, but the idiocy of that proposal is another matter (and i’m not referring to the dangers, but rather the alternatives which aren’t wind, hydro or solar)