Posted on 10/06/2025 11:31:48 PM PDT by delta7
OpenAI CEO Sam Altman is doing what he does best: warning of AI-related doom that he doesn’t sound all that pressed about.
Last week, while touring one of OpenAI’s mammoth data centers being built in Abilene, Texas, Altman had a pretty casual outlook when he was asked about a bubble that could financially decimate the entire industry.
“Between the ten years we’ve already been operating and the many decades ahead of us, there will be booms and busts,” Altman ruminated, as quoted by The Associated Press. “People will overinvest and lose money, and underinvest and lose a lot of revenue.”
“We’ll make some dumb capital allocations,” he conceded — but assured that “over the arc that we have to plan over, we are confident that this technology will drive a new wave of unprecedented economic growth.”
It’s the kind of blasé, bromidic talk you’d expect to hear from a coach of a sports team that’s on a historic losing streak. Oh yes, “there’ll be ups and downs” — and your eyes glaze over. You forgive them, though, because, well, what else are they going to say? That everything’s going down the toilet?
But Altman commands a half-trillion-dollar startup that’s the tip of the spear for an out-of-control AI gold rush. Pretty much the entire world economy is tangled up in the hundreds of billions of investment being poured into the industry.
Illustrating just how all-devouring AI is, here’s this worrying statistic provided by the Wall Street Journal: in the US, capital expenditures for AI contributed more to growth in the economy in the past two quarters than all of consumer spending, according to Neil Dutta, head of economic research at Renaissance Macro Research, citing data from the Bureau of Economic Analysis.
So it’s hardly an exaggeration to suggest that if the AI bubble bursts, it could take the whole economy down with it.
And the omens are everywhere if you look for them — the most looming being the fact that the biggest AI firms haven’t demonstrated that they can turn a profit with the tech that everyone is banking billions of dollars on revolutionizing productivity.
But nonchalant doomsaying is Altman’s specialty. And it’s not the first time that he’s warned about an AI bubble. In August, he outright said we were in one.
“When bubbles happen, smart people get overexcited about a kernel of truth,” Altman told reporters, as quoted by The Verge. “Are we in a phase where investors as a whole are overexcited about AI? My opinion is yes.”
He’s also warned for years that AI will wipe out entire categories of jobs, cripple society by drowning it in misinformation, or perhaps cause a Terminator-style apocalypse that lays waste to the human race.
Of course, when Altman makes these prophecies about AI, he’s also hyping it up. And perhaps part of why he can seem so nonplussed about impending catastrophes, financial or existential, is that he’s convinced that his company will be the one that comes out the other side stronger.
Right after warning of a bubble in August, Altman acknowledged that “someone” will lose a “phenomenal amount of money,” but that, of course, “we don’t know who.
![]() |
Click here: to donate by Credit Card Or here: to donate by PayPal Or by mail to: Free Republic, LLC - PO Box 9771 - Fresno, CA 93794 Thank you very much and God bless you. |
Sam Altman always has a hidden agenda.
In this case, I would guess that he is trying to suppress venture capital funding for up-and-coming startups that are potential competitors.
But maybe his purpose is something else. I just know not to take him at face value, ever.
But maybe his purpose is something else.
————
Everything in today’s world is Deception. I know little about Altman, but do know this AI thingy creeps me out….strange how it is “ blossoming “ when the world’s financial systems are imploding ( debt) and societies are fracturing.
AI isn’t going to correct our troubles, but most likely used to increase them.
Bkmk
That is real money being spent. And much of it appears to be money spent on bitcoin mining. As long as everybody who owns bitcoin believes in its inherent value, the value is maintained. Sort of like the collector price for a 1965 Mustang or a Honus Wagner baseball card.
If/when the air pops out of the bitcoin balloon, there will be huge capital losses taken that would affect income tax collections. Even if bitcoin continues to hold value, some of the data centers will fail for various reasons and the capital losses will appear.
But of course, MAYBE the AI craze is not like the dotcom bubble, or pet rocks.
This whole AI topic is reminding of the ‘dotcom’ bust, one big difference though. The dotcom explosion didn’t cause tens of thousands of acres bulldozed and just left there, water supplies depleted, enormous electric bills for residents, and other countless billions lost by big companies similar to EV investments, the list goes on.
I use ChatGTP at work for quick reference. Twice, it has answered with incorrect info, even incorrect within the info it provided.
I asked for a total x from 2020 to date, and then a yearly breakdown. The breakdown didn’t add up to the total. When I asked about the error, it kept coming back saying “you are right”, then give me the wrong total again.
Initially AI tools are making deception easier, however soon everyone will have an AI BS detector running in their AI bodycam that will indicate deception levels. That will be the end of leftism as we know it. The Demorats cannot survive without BS. AI bodycams might be the end of rampant corruption as well, the precursor of communism.
AI is not crypto. There are some overlaps but they are completely different bubbles, if you will.
I wonder if the new H1-B policy is supported by AI optimism?
Let’s hope so.
“capital expenditures for AI contributed more to growth in the economy in the past two quarters than all of consumer spending”
Economy grows, people have less to do.
The new H1B policy is a bait and switch. I don’t think it will reduce it all. It was supposed to be $100k a year, now it’s a one time fee. It was supposed to be immediate now it’s 3 years down the road. All these articles written by Indian rags are gaslighting us into thinking it’s actually gonna do anything to stop the flood of sub-continent trash.
I predict there will be a temporary window where AI is great at Forex trading, making easy riches for the people who figure it out. That window may have already passed.
It’s vaporware built on VC money aka debt. Huge debt for datacenter hardware with the expectation of payoff somewhere down the line. Just like the .com bubble madness -and crash. Throw money at some computer tech believing it has a potential for future profitability. Like .coms, etc. Even if the company makes money it has to pay off its VC and the massive money spent on datacenters and power.
NOPE
Well, I asked AI Grok: “How much has American companies spent on AI and how much revenue has it brought in”
“In summary, US companies spent ~$280B combined on AI capex and private investment in 2024 alone, but generated ~$146B in revenue—highlighting high upfront costs with returns still emerging. Projections for 2025 show spending accelerating faster than revenue, potentially straining balance sheets unless adoption scales.”
So, there is another finite resource that has to be taken into consideration: Capital Expenditure (capex). These AI companies are spending $280B but only bringing in $146B in revenue. That is not sustainable in the long run.
everyone will have an AI BS detector running in their AI bodycam that will indicate deception levels.
—————-
Perhaps, but…..
“Men are nearly always willing to believe only what they wish”.
-Julius Caesar
Human nature shows us man commonly forms his perceptions to align with his desires and not evidence or reason. Emotions hinder reasoning. Not many humans have been given the Gift of Discernment.
And here I was getting excited. I was on the top of Mt Monadnock Sunday. There were packs of them.
But AI will only ever be as good as those who compile its language models. When the work is done by domestic "little pinks" under the direction of a Sam Altman, the user just gets whatever Left-biased answers Reddit or Wikipedia would provide. If an AI is actually allowed to notice and analyze things, it comes up with all sorts of uncomfortable answers.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.