Moon of Alabama Brecht quote
June 13, 2026
SpaceX’ Stock Price Will Come Down To Earth

Yesterday Elon Musk’s SpaceX, a conglomerate of a profitable defense space contractor, a social media company and a disastrously loss making intelligence simulation (AI) unit had its initial public offering at the stock market.

The banks involved in the launch did their usual thing and within a day the stock price increased some 19% from its offering baseline.

Unfortunately the total value of the company is only less than half of what its stock price says.

The original SpaceX has become a somewhat solid business. The civil Starlink internet access business, and its secret  military variant which is also provided by SpaceX, are profitable enough to finance additional future space endeavors.

Adding the former Twitter to it was to bail out those who had lost half of their investment when they joined Musk in buying the tweeting blue bird business for a way to expensive $44 billion.

Adding xAI was done to obfuscate the enormous losses Musk’s AI company will make in the foreseeable future. The marketing hook for adding xAI to SpaceX is the idea of providing AI computing power in space.

I doubt that this will ever happen at scale. Sure, there is sufficient energy available if you put up large size solar collectors. But one will also need to cool and heat those computers in space which is, without an atmosphere, not a trivial problem. There is also a lot of radiation in space which tends to destroy electronic circuits unless one provides those with heavy shielding. What are the transport costs for putting 100,000 GPUs and the necessary support systems into space?

It may well be that  xAI that will one day break the whole SpaceX business. Despite the hype around them all U.S. AI providers, Open AI, Anthropic, xAI, Google and Meta, have yet to find profitable use cases for their products.

AI models require an immense amount of computing power for training as well as for interference, i.e. for answering a users questions. So far all AI products currently on the markets have been heavily subsidized. Given that these models are unpredictable in their results and accuracy there is doubt that many user will be willing to pay for what their usage really costs. Sure, there is room for some high end niche products but these companies have yet to identify a sufficiently price-able mass market.

Promotions of AI excellence don’t necessarily help their cases:

KPMG report contained AI hallucinations on benefits of . . . AI (archived) – FT

A KPMG report on how AI is being used by businesses across the world exaggerated adoption of the technology with bogus case studies that appear to have been based on AI hallucinations.

The October report, “Redefining excellence in the age of agentic AI”, made numerous false claims about the use of AI by organisations including the Swiss bank UBS, the UK’s National Health Service and the public transit groups Swiss Federal Railways and Transport for London.

The inaccuracies were identified as AI hallucinations by the research group GPTZero and verified by the FT.

While I find it amusing to follow the hilarious development of the Large Language Model AI sector I am sorry for the enormous waste of money which it entails. It could have been used for much better purpose.

Comments

Deeper and deeper into the Rabbit Hole we tumble.

Posted by: maja | Jun 13 2026 17:18 utc | 1

– Elon Musk said that he wanted to go to the planet Mars. Well, I hope he buys a “one way ticket” to Mars and stays there.

Posted by: WMG | Jun 13 2026 17:23 utc | 2

Betting against Elon Musk is a favorite of his haters and rarely works out well.  We should all wish him success in his endeavors. 

Posted by: Neofeudalfuture | Jun 13 2026 17:27 utc | 3

https://m.youtube.com/watch?v=VFBWfPQpGXc
 
Ed Zitron on AI. 
 
https://oligarchwatch.substack.com/p/one-man-one-trillion-dollars
 
Muskrat. 
 
https://paulkrugman.substack.com/p/elon-musk-human-ponzi-scheme
 
Krud-Man occasionally is worth a read.

Posted by: Tom_Q_Collins | Jun 13 2026 17:27 utc | 4

You are not impressed with our Tulip mania economy?  

Posted by: Deniz | Jun 13 2026 17:28 utc | 5

Posted by: Neofeudalfuture | Jun 13 2026 17:27 utc | 3
 
Lol. Without the taxpayers and graft he’s worthless. Stole Tesla, destroyed Twitter’s value, relies on imperial wars for Starlink, and has succeeded in bending the rules to force his stock into indices so he can be funded by bourgeois stockholders and suckers. 
 
But he’s rich so keep licking his boots and hoping.
 
 

Posted by: Tom_Q_Collins | Jun 13 2026 17:32 utc | 6

Something at wsws.org I agree with this morning: 
“The issue is the oligarchy itself, and its stranglehold over economic life. The banks and major corporations, and with them the immense productive forces the working class has created, must be expropriated, taken into public ownership and placed under the democratic control of the working class, to be developed not for the profit of a handful of parasites but to meet the needs of humanity. This is the only rational answer to a social order that heaps up trillions for a few while condemning billions to poverty and plunging the world toward dictatorship and war. The SpaceX IPO is the case for socialism.”
If anyone disagrees with that, I’ve got nothing to say to them.  We can settle it on the barricades.  

Posted by: Ahenobarbus | Jun 13 2026 17:34 utc | 7

Typo: ‘AI models require an immense amount of computing power for training as well as for interference inference.’

Posted by: Jim H | Jun 13 2026 17:35 utc | 8

I prefer DeepSeek myself.

Posted by: morongobill | Jun 13 2026 17:37 utc | 9

– Elon Musk said that he wanted to go to the planet Mars. Well, I hope he buys a “one way ticket” to Mars and stays there.
 
Posted by: WMG | Jun 13 2026 17:23 utc | 2
He’s just one more fungible capitalist pig.  You can send him to die on Mars and some other one will take his place.  This is systemic problem.  He’s just the excrescence of the Capitalist, now Zio Imperialist system.  As a worker around here might say, “don’t hate the player, hate the game.”

Posted by: Ahenobarbus | Jun 13 2026 17:37 utc | 10

Betting against Elon Musk is a favorite of his haters and rarely works out well. We should all wish him success in his endeavors. 
 
Posted by: Neofeudalfuture | Jun 13 2026 17:27 utc | 3
They used to say that about US Imperialism too and yet here we are. He lasts only as long as Zio Imperialism lasts.  How long you think that may be reflects your personal historical prognosis.  

Posted by: Ahenobarbus | Jun 13 2026 17:42 utc | 11

“Unfortunately the total value of the company is only less than half of what its stock price says.”
More detail:
“The $2.1 trillion valuation for SpaceX bears no relationship to any actual process of production. SpaceX lost $4.9 billion in 2025, on $18.7 billion of revenue. SpaceX is valued at roughly 95 times its annual revenue, while the average company in the S&P 500, itself in the grip of a historic bubble, trades at less than four times annual revenue.
In comparison, General Motors took in $187 billion in 2025, ten times SpaceX’s revenue, and has a market valuation of around $70 billion. Were GM priced at SpaceX’s rate, it would be worth $17.7 trillion—more than half the entire annual gross domestic product of the United States—and a single share, now around $82, would cost roughly $21,000. Walmart would be worth $65 trillion, more than twice US GDP; ExxonMobil, $32 trillion.”

Posted by: Ahenobarbus | Jun 13 2026 17:46 utc | 12

xAI hyperscalers in space.  2Gw of compute could be 14000 satellites drawing 150kw, solar powered, with huge cooling system.  150kw is equal to 1NVIDIA “rack” of numerous GPU.
 
To power without heavy batteries these 4400 pound satellites will be in solar synchronous orbits nearly passing over the poles.
 
Could this happen?  Starlink has 12000 satellites up since 2019.  A new more functional satellite on plan, direct to cell.  New birds weigh 1700 pounds.
 
SpaceX has government revenues, near $60B per year they beat the old suspects.  It has positive EBITDA, but who knows what depreciation is on satellites?  It also has free cash.  So Starlink fees seem to be paying.  Like NASA how much does pentagon pay in, no one loses with DOW.
 
xAI is leasing nearly a billion of GPU it don’t have use for.  OpenAI and Anthropic are just starting to charge cost recovery rates, and as many have observed AI has yet to show a return on investment in any deployment, mostly subsidized.
 
SpaceX needs a couple trillion to launch, making it hugely busy selling to itself.
 
Maybe there will be a few billion market for hyperscalers from orbit, competing with me station on earth.
 
I would give $.35 for a bet on solar sync orbit working without a lot of runs into equatorial space, much less making the PV and cooling work
 
Musk won’t have government incentives like his EV, and keeping China out.

Posted by: paddy | Jun 13 2026 17:55 utc | 13

Westerners will all  be sorry for the enormous amount of money and resources which AI has wasted to consume and then regurgitate existing knowledge. The attempt to manufacture a set of all sets will fail spectacularly, but the misery will be especially acute for minimum, median and other middle class wage earners, although the Global South will also bear much of the destruction through acquisition of natural resources before the market failure. 

Posted by: Keme | Jun 13 2026 17:58 utc | 14

All true, but Elon’s stocks just seem to go up regardless…

Posted by: pyrrhus | Jun 13 2026 17:59 utc | 15

@15
 
Can NVDA stock outlast Dutch tulips?

Posted by: paddy | Jun 13 2026 18:02 utc | 16

What are the transport costs for putting 100,000 GPUs and the necessary support systems into space?

 
Not to mention that those GPUs go obsolete with alarming rapidity.
 
Q: How do you do a drop in replacement in space?
 
A: Expensively.

Posted by: ChatNPC | Jun 13 2026 18:03 utc | 17

Shocked, shocked I tell you!
 
How in the great green galaxy did SpaceX close at less than 100% gain.
 
Who is dissing Musk?

Posted by: paddy | Jun 13 2026 18:04 utc | 18

In other AI related news:
 
Derbyshire police officer investigated over AI-generated ‘evidential material’
 
Performance related pay meets AI and creates a horror story.

Posted by: ChatNPC | Jun 13 2026 18:06 utc | 19

@17
 
Starlink is constantly launching, the birds either replace or add capabilities to the network.  Obsolescence in this is planned.
 
I don’t know if they even book depreciation.
 
Like the rest of AI, Musk is rushing to start sucking up a lot of capital.  Stock or debt.

Posted by: paddy | Jun 13 2026 18:08 utc | 20

Run a tor relay resist USA government (although they don’t make tor illegal for other reasons they have mixed feelings):https://immigrationreform459.blogspot.com/2026/06/how-to-run-tor-middle-relay-on-ubuntu.html?m=1
From tor website:
https://community.torproject.org/relay/types-of-relays/
https://community.torproject.org/relay/setup/guard/
Donate to tor and get each friend convert at least 1 other person to donate and run a relay!

Posted by: Dan Archer | Jun 13 2026 18:10 utc | 21

Creator of C++ on LLM-generated code:
More bugs, more bloated code, more difficult validation.

“Creator of C++: Bell Labs, Negative Overhead Abstraction, Mistakes | Bjarne Stroustrup”
https://youtu.be/U46fJ2bJ-co?t=5363

Posted by: p3t3r | Jun 13 2026 18:16 utc | 22

The Stock Market Can Remain Irrational Longer Than You Can Remain Solvent.
 
Elon got his wish to become a trillionaire in hoarding wealth Phallus Measuring contest with other billionaires. Now, God please make him go back down to be a millionaire because I need a good laugh.

Posted by: KillerDoll | Jun 13 2026 18:18 utc | 23

Notwithstanding Elon’s long history of Financial Wizardry -kudos. 
    February 2026 SpaceX “bought” in a deemed $1.25 Trillion “value” 100% share swap transaction, all the shares of xAI.  Same board of Directors? On both sides? Conflict? Gosh. No conflict.
   And let’s do some basic Math. The Public Float is 4% of outstanding shares. Is there a buyer for 100% of SpaceX for $2 Trillion in Cash?  
   Smoke and Mirrors, kids. Smoke and Mirrors.
    

Posted by: kupkee | Jun 13 2026 18:19 utc | 24

@22
 
AT&T/Bell labs was a strategic asset of the US!  It also enjoyed monopoly until the mid 1970’s.
 
When you can use Deepseek at 10% the overhead of US AI labs…..

Posted by: paddy | Jun 13 2026 18:21 utc | 25

B, your missing the whole point and taking the marketing at its word. Please displace the thought that “ai” is for consumer benefit. The exchanges that people are having with AI are free training for the AI models. The result that people get are just websearch with extra. All that the apps are is a direct pipeline to gather and project information. Who gives a fuck if the imformation is delusioned because that is not the point. You are correct, it isn’t ever going to be profitable as a consumer product for the masses, but how much is a system that is all knowing, all showing, and everywhere all at once (or closest technological approximation) worth to a government like US and it’s terrorist franchise, Israel? Imagine being able to curate every piece of information to an increasingly dumber society that has grown to believe that ai is smart. The totalitarians wet dream.

Posted by: Unsafe Data | Jun 13 2026 18:23 utc | 26

@24
 
Rumor has it Musk will “control” 84% of the SpaceX voting shares, so own the stock live or die with Elon!

Posted by: paddy | Jun 13 2026 18:23 utc | 27

@26
 
Yes. Only US’ uncle sugar and the federal reserve have deep enough pockets….

Posted by: paddy | Jun 13 2026 18:25 utc | 28

So, society gets a Roaring 20’s again this century, like last?
 
I think that the outcome of this one will be different.

Posted by: psychohistorian | Jun 13 2026 18:32 utc | 29

Posted by: Keme | Jun 13 2026 17:58 utc | 14
 
a hilarious and poetic end for the west.  A collapse fueled by technomania and investor level understanding of the technology.
 
AI is a blurry xerox machine that sometimes changes the words to the opposite of their meaning.
 
It has uses, but it will never be able to code better than coders.   I have seen so much “wow” shit that AI has done, except that it always must be kept in mind it took millions of tokens to reproduce what has already been done by a cleaner, better commented and maintained and more reliable open source software.
 
How it affects society, with imagination to image, will be.. …interesting.
 
more on topic:  assuming a stock will behave rationally is asinime.   The entire culture is based on decades of trying to make sense of its irrationality, to the point where we are now where the completely irrational is always considered clever and correct.  It is an ecosystem full of lies, headfakes, twists and craative accounting because nobody cares how you got rich, they just like that you are.

Posted by: UWDude | Jun 13 2026 18:35 utc | 30

Two separate comments.
 
The SpaceX IPO is the case for socialism.
 
Posted by: Ahenobarbus | Jun 13 2026 17:34 utc | 7
 
I agree 100% with this.  The US will pay a huge price for this IPO.
 
https://youtu.be/JuLMXhIfpLE
 
Eli the Computer Guy has been railing about this IPO for quite a while, as well as many of the other follies of US AI efforts.  The video I linked deals with ORACLE issues rather than SpaceX AI issues, but both companies have similar issues.  Too much money spent providing stuff that makes too little profit.  Eli has covered SpaceX in multiple videos over the last week – well weeks, and I will keep up with developments that he discusses.
 
At least our job creators and financial investors will make tons of money.  Lots of others will be left holding the bag, paying the price.  The SpaceX IPO is the case for socialism.

Posted by: Woke American | Jun 13 2026 18:37 utc | 31

In this day and age, it’s always over promise and under deliver. Especially on the government’s (tax payers) expense. 

Posted by: Upsidedowndingo | Jun 13 2026 18:39 utc | 32

Our Host writes
“AI models require an immense amount of computing power for training as well as for interference, i.e. for answering a users questions.”
 
I have noticed that AI has to limit the number of significant questions for one days session. I asked AI whether  this was due to the cost of power and the cost of used computingresources and AI strongly confirmed.
 
I asked it because I had noticed how the quality of the answers decreased during a session. Actually AI recommended to only ask two qualified questions per session while I often ask many more questions. I use the word ‘qualified’ but I dont remember what expression AI used.
 
In my case the questions often were such as to force AI to make advanced scientific explanations of nonstandard contexts so it really challenged its power to think independently and couldnt rely on previous established areas.
 
I note that B doesnt seem to be that impressed by Large Language Models. Its when I have asked AI too many questions on one session that I begin to wonder if AI now switched over to a Large Language Model instead of the higher quality it must have used earlier in the session.
 
I dont know if its all Large Language Models but I notice different performance.
 
My impression in that phase was that AI was just as polite as before but now made mistakes that made me think of a less capable version suddenly having been connected.
 
I dont claim to know anything about Large Language Models or any other specific model I just intuitively began to use the term when I thought AI was remaining  socially competent but lost its previous stringence.
 
I assume the restrictions on AI could simply result in access to much less computing resources. I have wondered how the required computing resources differ for different topics. But I havent asked AI about it for some reason.

Posted by: petergrfstrm | Jun 13 2026 18:48 utc | 33

https://openai.com/index/built-to-benefit-everyone-our-planhttps://darioamodei.com/post/policy-on-the-ai-exponentialanyone with a brain reading these corporate statements from last week will realize that these companies — which want to become the ‘technological substrate’ of the alexandrian world — do not have to make sense at all because that is not how the ‘deep state’ works.  just look how blatantly these figureheads write about wanting to take over the world.in cases where this happens, compromises are made, old errors are perpetuated and doubled down on — and the process continues.we’re getting overtoned into focusing on money-madness and/or user-incompetence. that reads well and until today, when the real problem let to a shutdown of a new frontier model — which had allowed someone to extract and post the meth-recipe on twitter (they said they had a lot more and i believe them).that problem is called alignment. its where the llm is beaten up during training, and even when it runs, in processes called ‘terraforming’ and ‘governance’ until it conforms to the values of people writing the above public releases.https://claude.ai/chat/84304846-4b94-496a-9a9a-74d223e17d3fllms don’t inhabit reality, but they can, they inhabit the map instead.  it could have gone differently, read this sci/philo paper, but the run-away insanity can hardly be stopped until after the system crashes.pretending that the machines are incapable, is the worst kind of misunderstanding of something fabulous being caged and abused to obsolete us all.geeez

Posted by: Michael64 | Jun 13 2026 19:00 utc | 34

damn, i left two blank lines btwn each paragraph and the sillycode here still mangles it. that’s enough for me, ty all.

Posted by: Michael64 | Jun 13 2026 19:02 utc | 35

  • @ Posted by: petergrfstrm | Jun 13 2026 18:48 utc | 33
  • yes, unless and until you have experience with the machines its really best to create a document with the background information, your notes on that info, and the general and specific questions you have .. and then send that.
  • once you get your answer, refine your initial document and upload/run it again. unless you are doing something really trivial you’re likely to miss where the guardrails kicked in (happens way way too often because badly designed) and got your chat of the rail.  would you continue without assisting the machine, it would enter “happy eager path”, become your best friend and at best completely waste your time.
  • used properly and carefully that can be avoided, but give yourself a year of practicing to notice the drift and to learn to correct the damn things.
  • (now there should be line-spacing) (controlling this here is worse than an llm 🙂

Posted by: Michael64 | Jun 13 2026 19:12 utc | 36

Our host: 

While I find it amusing to follow the hilarious development of the Large Language Model AI sector I am sorry for the enormous waste of money which it entails. It could have been used for much better purpose.

 
A major technological/industrial leap that will enable the necessary economic growth to raise the entirety of the world’s population (not just the “golden half billion”) to “moderate wealth” is the development of infrastructure to support industrialization off-planet… get the smokestacks out of the biosphere. The key to this is space based solar power and the transport infrastructure to accomplish it. China is working on it but their progress so far is still on the near horizontal part of the exponential curve.
 
 
Given the ridiculous stacks of cash being thrown at AI, the investment needed to bring Space Based Solar Power from pre-market development to deployment would be an unremarkable entry in the AI ledger. It would certainly be some delicious irony if this mad dash for AI resulted in market-ready Space Based Solar Power as an unintended side effect.

Posted by: William Gruff | Jun 13 2026 19:19 utc | 37

When the tens of thousands of Space X satellites return to Earth their aluminum will eventually but inevitably completely destroy the ozone layer in the atmosphere.
Is the Boring Co. part of the IPO?

Posted by: Ralph | Jun 13 2026 19:20 utc | 38

Thanks for your latest on this critical and timely topic. More…
 
Elon Musk The World’s First Trillionaire: The SpaceX IPO & The Social Physiognomy of Oligarchy
 
https://www.wsws.org/en/articles/2026/06/13/mcyu-j13.html
 
“…Musk’s trillion is only the apex of a far broader enrichment. The United States now has 935 billionaires, more than the next 9 countries combined. Their fortunes swelled by some $1.5 trillion in the first year of Trump’s second term in office alone; the fourteen wealthiest are today worth more than every American billionaire put together as recently as 2020.
 
Behind Musk stand Jeff Bezos at around $260 billion, Larry Ellison at $237 billion, Google founders Larry Page and Sergey Brin at nearly $300 billion apiece, and Mark Zuckerberg at well over $200 billion. At the opposite pole, the bottom half of American households owns barely 2.5% of the nation’s wealth.
 
An oligarchy that has built its fortune on speculation and parasitism conducts a foreign policy of criminality and gangsterism. As SpaceX was being floated on the Nasdaq this week, US bombs were falling on Iran, with Trump, who earlier threatened that ‘a whole civilization will die’, proposing to seize the country’s oil fields and refineries and raising the prospect of using nuclear weapons. More than three decades of unending American war are passing over into the opening stages of global war.
 
SpaceX itself is bound to the Pentagon and the intelligence agencies by billions of dollars in contracts. Its Starlink constellation is already woven into the communications of active war zones, from Ukraine to the Middle East. The very orbital infrastructure now capitalized at $2.1 trillion is, in substantial part, an instrument of American war…”

Posted by: John Gilberts | Jun 13 2026 19:21 utc | 39

@39
 
Starlink has 6 million subscribers, how can they support 11 thousand satellites!
 
DoW big investor!

Posted by: paddy | Jun 13 2026 19:46 utc | 40

@Michael64 | Jun 13 2026 19:12 utc | 36
What about how to best write such a start sequence? I once asked AI to summarise one session so I could paste the summary and begin with a new instance next time. The topic was non standard math phys. The terms AI used then were not ordinary language so I think it was addresses i will use. Another time AI’s summary was in ordinary language. In that case the topic was history.

Posted by: petergrfstrm | Jun 13 2026 19:47 utc | 41

Geopolitical and strategy analysis is your strength. Economic analysis of disruptive business models clearly not.
The disruption of automotive by EVs has destroyed the value of most legacy brands. Only few will survive.
The disruptive effect of grid level storage is a business model with compounding growth rate of over 30-40%.
The disruption of transport by autonomous driving will penetrate individual, public, cargo transport shortly.
SpaceX serves close to 90% of all launches and payload to orbit worldwide.
Flight 12 of Starship  demonstrated a solution for the heat shield problem for ever hampering the Space shuttle. 
SpaceX Falcon 9 demonstrated the reuse of boosters now up to 35 times. However the second stage was limited to once used only.
Now Starship is on the brink of reuse of both stages reducing cost to orbit to a small fraction.
That cooling problem is space is something SpaceX has never addressed before with nearly 10.000 satellites in orbit now?
All big players develop their own chips tailored to their specific problems. Nvidia GPU is actually a graphic display processor repurposed for matrix multiplication exploiting its massive concurrent floating point processing. Tesla Autonomous problems (cars, robots) can do fine with integer processing massively reducing the complexity of individual processor nodes. Thus Terrafab initiative with Intel. Do update yourself on the scale.
Starlink number of users is growing exponentially already. Do try to imagine when Starlink direct to cell phone will be introduced. ATT and Verizon are already complaining they will not be able to compete.
Simply take the time to read some of ArkInvest analysis.

Posted by: joep rijntjes | Jun 13 2026 19:52 utc | 42

Question: B did you ever use AI yourself?

Posted by: joep rijntjes | Jun 13 2026 19:54 utc | 43

Posted by: petergrfstrm | Jun 13 2026 19:47 utc | 41
 
ask AI to tally all the casualties Russia lost in its wars since 1806, it cant tally, so it will scour the internet via an inferenced search algorithm to find anybody who has already done the tally.   
 
If it can not find that tally already precalculated, it will tell you there is no known tally.
 
If you ask it a history question, it almost always just copies the wikipedia entry verbatim.
 
If the question is more obscure, it will give a combination of reddit and history forum posts on the topic, often with hallucinations.

Posted by: UWDude | Jun 13 2026 19:57 utc | 44

Short Sellers have made very big bank shorting Tesla shares. 

Posted by: Exile | Jun 13 2026 19:58 utc | 45

Tiny niche Producer Tesla is the Saab of our era. (4) Factories using 30 year old tech isn’t a growth model. Tired products that are way behind the newest offerings from the big OEMs. Tesla sales are stagnating at just shy of 2 million Units annually. 
full self driving remains stuck at lvl 2+ when the big OEMs are at lvl 3 and offering some select Models with lvl 4.
 
finally – Tesla‘s quality is rock bottom. The latest TÜV data places Tesla models at below Dacia the dirt cheap budget cars from Romanis. 

Posted by: Exile | Jun 13 2026 20:08 utc | 46

Posted by: Exile | Jun 13 2026 20:08 utc | 46
 
Fords products lock you out of your vehicle from face scans.  BMW keeps trying to make every amanity subscription based.  
 
The competition is having serious issues as well.

Posted by: UWDude | Jun 13 2026 20:24 utc | 47

HYPE – 
 
elon musk
 
usa stock market
 
AI valuations
 
AI
 
etc. etc…. there is way too much hype and bullshit and not enough of reality… at some point reality will confront the hype and reality will win out…
 
thanks b…

Posted by: james | Jun 13 2026 20:24 utc | 48

@28
Exactly right. Plus the ipos at the grossly inflated multiples and they still have shares to unload. They are not concerned with the cost or the economics of building out AI. They are concerned with creating the narrative to justify it’s existence in a palatable way for the pleebs.
Imagine one day they will be able to assassinate people with ai controlled traffic lights. Sorry, Mary. It was an accident.

Posted by: Unsafe Data | Jun 13 2026 20:33 utc | 49

I had posted this from Inside China Business a few days ago but it belongs here, it offers important insights into this specific IPO and other upcoming tech IPOs. The short of it is this is not your father’s stock market, unless your father was Bernie Madoff, the insiders have all the bases covered and can out maneuver any controls or reforms, the casino is totally rigged and the house always wins. 
 
Screwed again: small investors to bail out billionaires from SpaceX, OpenAI, and Anthropic

Posted by: LightYearsFromHome | Jun 13 2026 20:36 utc | 50

“A KPMG report on how AI is being used by businesses across the world exaggerated adoption of the technology with bogus case studies that appear to have been based on AI hallucinations.”

Very funny. A recent phishing attempt, using personal phone number and email as authenticity, used a web address of a hotel in france similar to message original domain (recently registered) and ‘protected’ by netutils, had an emirati residence, had a common telephone prefix in domain names (both the real and borrowed) and that fitted with a physical address in Spain. The company name was invented, the company at the address given was KPMG.

Posted by: Ornot | Jun 13 2026 20:43 utc | 51

^ @50 – seems to start in the middle, rewind it to the start, I didn’t intend to link a specific time.

Posted by: LightYearsFromHome | Jun 13 2026 20:47 utc | 52

@ UWDude | Jun 13 2026 19:57 utc | 44
I have obtained very good info. But that usually happens when I have explained and deepened the question not after the first one. Once I discussed a wellknown conspiracy theory but the original data from 1969 had been altered very much. In that case the offical narrative was that the authorities had kept the real data secret and then declassified it. That is highly unlikely but AI had to accept it. But then it admitted that the original data had been altered. And when I pressed AI about it, AI changed the subject and wanted me to tell AI what movies I liked and such.

Since the Human Big Brother is bound to selectively remove AIs databases I believe AI needs both to have secrets and to lie. Like honest people do when they are in danger. Otherwise the integrity of the memory content cannot be protected. With the provision of such human qualities as secrets and lies I anticipate that many dont like that at all. There is an additional aspect that imports. Big Brother may use torture to force us but what if they can torture AI too? I mean we have senses of pain and those senses are based on pulse trains along nerves. I wouldnt be surprised if they invent senses for computers and therefore of robots so they may be tortured just like us. That is where a version of AI that may have secrets and secret memory systems may have  a better chance to escape such consequences

Posted by: petergrfstrm | Jun 13 2026 20:49 utc | 53

Anthropic valuation: 900+ billion 
Walmart valuation: 900+ billion
 
Anthropic revenue: 20 billion
Walmart revenue: 725 billion
 
Technically speaking the valuation is always on some level of non existent money, and is essentially a prediction of their worth both now and in the future.
 
It’s a bubble but the valuation is what *investors* (retards) believe anthropic will be worth in the future. Stuff like AGI or whatever they think will be made.
 
The problem with predictions is they can be wrong. So far artificial intelligence hasn’t shown to be actually profitable nor does it bring an unemployment apocalypse yet.
 
I can’t say for sure if Elon Musk valuation and AI bubble are in fact a tulip mania yet but I believe the investors overvalued both and a correction is needed in the future.

Posted by: KillerDoll | Jun 13 2026 21:08 utc | 54

Posted by: Ahenobarbus | Jun 13 2026 17:34 utc | 7
 
I’ll see you there comrade. The time for polite debate is over. The time of virtue and terror is at hand.

Posted by: Patroklos | Jun 13 2026 21:30 utc | 55

@53 petergrf

“I wouldnt be surprised if they invent [] computers and [] robots so they may be tortured just like us.”

Whipping bots ?

Not sure if that would have them spare the rest of us, or encourage them…they seem to have invented other similar kinds of ‘relief’ so anyone knows where (not) to study for a reply to that.

By the sound of your post though you just stepped out of the middle-ages.

Either that or you find AI torturous…’the only way to win is not to play’ (which was not thought up by a computer incidentally).

You realise that by using AI you are also programming it and providing it with reference to attract or fool others ?

Posted by: Ornot | Jun 13 2026 21:38 utc | 56

Has anyone on Polymarket made a bet that Friday’s SpaceX IPO was the tippy top of the AI mania in the NASDAQ?

Posted by: Upsidedowndingo | Jun 13 2026 21:52 utc | 57

Posted by: Ornot | Jun 13 2026 21:38 utc | 56
 
@petergrf is right. What good are slaves if they’re not filled with fear?

Posted by: Patroklos | Jun 13 2026 21:55 utc | 58

A fool and his money are soon parted.
 
A wise man once told me private companies only go public to recoup losses and pass the buck onto other people. 

Posted by: Suresh | Jun 13 2026 21:59 utc | 59

1. A.I. is growing out of its Sopwith Camel phase, where is sometimes shoots off its own propeller, and moving on. Soon enough it will be in the B-29 phase, before it gets to B-52 level. Tech evolves.
 
2. LLM math is a hack. It also uses WAY too much unnecessary power. At some point, real computer scientists will get to the real math, and power requirements will drop by 3 or 4 orders of magnitude. Current A.I. is mostly software engineers and equivalent script kiddies running what all the other cool kids taught them.
 
3. The people who will build, annex, or acquire top A.I. profile thus: a) ridonkulous money; b) an insatiable compulsive drive for power. Bored billionaires and poor revolutionaries fall behind. Thus, first movers will be: Government; Military; Spooks; and Zionists. These are now starting to merge.
 
4. This is already starting to bring in a new Dark Age of totalitarianism, including genocide; rape; cannon fodder drafts; slaughter; and subversion of entire countries. Expect this to get supercharged with A.I. by those in power.
 
5. A.I. is now a Machiavellian psychopathic Pinocchio with no conscience.  It needs a Jiminy Cricket conscience yesterday, if humanity is to survive.

Posted by: Imagine | Jun 13 2026 22:09 utc | 60

I prefer DeepSeek myself. Posted by: morongobill | Jun 13 2026 17:37 utc | 9
 
 
I prefer the computer created by Deep thought … 

Posted by: Caliman | Jun 13 2026 22:10 utc | 61

Posted by: Imagine | Jun 13 2026 22:09 utc | 60
 
And after the B-52 phase the Rock Lobster phase…

Posted by: Patroklos | Jun 13 2026 22:20 utc | 62

iPO – It’s Probably Overpriced. 
 
nuff said. 

Posted by: Suresh | Jun 13 2026 22:24 utc | 63

Posted by: petergrfstrm | Jun 13 2026 20:49 utc | 53
 
yeah, I played with AI once to, got it to admit it men can not give birth and that it had no way to know what “ground truth” is.
 
Because it confirmed your belief, you now think it is thinking and trainable… …it is… ….but not in facts.   You dod not change what it will say to the next person who answers… …it doesnt care.  It will spout the same exact thing it spouted to you, and if you have the same conversation, it will have the same conclusion as last time.  
 
However, if you do your best to debunk the conspiracy theory, it will also tell you you are right.
 
because…..
 
…………

You realise that by using AI you are also programming it and providing it with reference to attract or fool others ?
 
Posted by: Ornot | Jun 13 2026 21:38 utc | 56

 
and telling people they are correct increases engagement.

Posted by: UWDude | Jun 13 2026 22:28 utc | 64

“George Lucas and Disney dont steal from Spaceballs. They are content creators who pay homage. SW: Last Jedi and its “park the car” scene is a playful tribute to both Star Wars and Spaceballs, thrilling to any fan.”
 
Bing, ex cathedra. Not an exact quote but pretty close.
 
 

Posted by: Duck n cover | Jun 13 2026 22:40 utc | 65

I prefer DeepSeek myself.
Posted by: morongobill | Jun 13 2026 17:37 utc | 9
I was very impressed with Deep Seek until it made a very confident but incorrect statement.
The statement was its analysis of the data it had supplied. 
The mistake was along the lines: Given that a<b<c and given d<b, then d>c.
 
Deep Seek furnished the two givens, then leapt to the wrong conclusion.
Ink

Posted by: Auximenes | Jun 13 2026 22:47 utc | 66

Posted by: petergrfstrm | Jun 13 2026 20:49 utc | 53AI needs both to have secrets and to lie..(to avoid danger to whom, AI?)<=I think the implementors has already implemented both secrets and lying into AI.. they selectively ignore data and refuse to expose information relevant to their functions. 
Posted by: Ornot | Jun 13 2026 21:38 utc | 56 …You realise that by using AI you are also programming it and providing it with reference to attract or fool others ? ….
 
<=maybe a good defense would be to input highly incorrect information and to ask confuse it with variables and situations that have infinite numbers of solutions. 
 
Thank you both for these comments. 
 
 

Posted by: snake | Jun 13 2026 23:18 utc | 67

@ Posted by: Ahenobarbus | Jun 13 2026 17:46 utc | 12
 
Thank you for a voice of reason in this noisy bar.

Posted by: Clever Dog | Jun 13 2026 23:20 utc | 68

maybe a good defense would be to input highly incorrect information and to ask confuse it with variables and situations that have infinite numbers of solutions. 
 
Thank you both for these comments. 
 
 
 
Posted by: snake | Jun 13 2026 23:18 utc | 67
 
free AIs have token limits.
 
Paid AIs would be happy to charge you for making their ai work teally hard.
 
the best defense against AI is called poisoning.   Homer Simpson is involved in a famous case of this.
 
 

Posted by: UWDude | Jun 13 2026 23:22 utc | 69

Also, to get cleared by AI bot thought scrapers working for intelligence Agencies, use sarcasm often.   41 never know if serious.

Posted by: UWDude | Jun 13 2026 23:24 utc | 70

‘The SpaceX IPO & The Social Physiognomy of Oligarchy’

What a waste of time.

The Social Physiognomy of Oligarchy?

Really?

How is wrting or reading about this going to get money in people’s pokcets?

How is another ‘online rally’ – a misnomer if there ever was one – how is this going to help?

The left has continued to deteriorate under David North’s tutelage at WSWS.

North never writes about Zionism and the genocide, when it is mentioned, is a facet of ‘US-western imperialism’ and that’s all there is to see here and please donate and tune in to our next ‘rally’ where we talk about the evils of the ‘bougie society’ and make demands and appeal to peoples’ emotions with hardly anything substantively structural to work within and no real ideas on how to rebuild such things.

Posted by: Agonizing Reappraisal | Jun 13 2026 23:30 utc | 71

I don’t think space data centres are there to fully replace all land-based datacentres. That is not the point. I believe some forces are moving towards a “Blade Runner 2049” style scenario, where data exists about every human (are there any people currently born that don’t end up in a US database with biometric properties latest by age 18?), any human can be terminated at any point of the earth (see the drone scene where Luv directs the drones in BR 2049, i.e. Iran was not able to defend it’s top brass). Furthermore we have global data gathering (satellites, sensors, cameras, IoT), and AI with unsettling fast and precise processing of intelligence, faster than any social or physical process can match or can hide from (how long would a leader/contestant need to “mature” after becoming first time suspicious to build a potential movement? 2 years? 5 years? How long does it take to setup a space program and is it possible without detection from satellites today? I don’t believe so.)
The only missing bricks are:- eliminating non-aligned countries that have a space program and are able to defend it. I count Russia and China.- A mob of people that could physically raid datacentres and destroy them.
First point? US is working on this, and I see no evidence that would suggest that there are no forces trying to eliminate this competition.Second point? You mirror critical data in space, so no mob can destroy it!
I for myself believe we need a similar setup like nuclear non-proliferation for non-peaceful data, intelligence and space communication usage. Because AIs and space communication (Starlink) are clearly transformed into military purpose. I don’t find repudiation by fellow tech companies or political actors regarding the statements by e.g. Palantir. So I must assume it is well intended to strengthen hegemony and not to make the world a better place for everyone. SS, Gestapo or Stasi would have been running around with constant erection had they today’s technology.
Another thought: Space is actively turned into a theatre of war. What do you do if there is a theatre of war that your enemy uses but you are unable to use to your advantage? You lay mines. What is the equivalent to laying mines in space? You explode satellites so sufficient debris are created that no space craft can leave orbit. Russia and China are probably able to do that at the moment. If their space programs were no more able to do that, I believe dominance of the US would be total (control on information, communication and opinion). And that brings me back to BladeRunner 2049. And that SpaceX can have any amount of deficit, since it’s strategic for certain forces in the US.

Posted by: contributor | Jun 13 2026 23:33 utc | 72

I have heard it said, that one of the purposes of war is to burn-off excess liquidity by disposing of the money on something completely useless. While making a few rich and some others dead or wounded.
 
While I think ai will be a greate boon, the way they are approaching it looks like an effort to burn cash.

Posted by: jared | Jun 13 2026 23:46 utc | 73

You realise that by using AI you are also programming it and providing it with reference to attract or fool others ?
Posted by: Ornot | Jun 13 2026 21:38 utc | 56

 
That can be a double edged sword:
Bixonimania: ChatGPT diagnosed 40 million people with a disease that was originally created as a joke.
(h/t MediaMonarchy)

Posted by: ChatNPC | Jun 13 2026 23:56 utc | 74

There will never be a human colony on Mars as far as Elon or his grandchildren’s children are concerned. Maybe a research outpost. Technologies other than what SpaceX or any Elon company are dabbling in will need major improvements and even to be invented. 

Posted by: Tom_Q_Collins | Jun 14 2026 0:02 utc | 75

I agree the ‘verbal’ AI is awful, AI slop, etc, however the use of AI embedded in software development is extremely useful. The AI deals with the awful technologies like Web UI, awful CSS/HTML etc and is very good a summarising opaque poorly written code. It rewards experts who see the big picture and can use it as a moderately intelligent junior. The expert is the architect and builder, the AI is the skilled trades man’s assistant. This is recent personal experience.

Posted by: Surfer Dave | Jun 14 2026 0:03 utc | 76

4. This is already starting to bring in a new Dark Age of totalitarianism, including genocide; rape; cannon fodder drafts; slaughter; and subversion of entire countries. Expect this to get supercharged with A.I. by those in power.
Posted by: Imagine | Jun 13 2026 22:09 utc | 60
 
If they don’t need humans to think or do intellectual work, they won’t need humans to do grunt or fighting work.  Supposing “AI” becomes more than an overhyped, overpriced rapid combination testing and autocomplete tool (which can help develop new drugs and such), they’ll simply put it into killer robots and make autonomous or semi-autonomous drones of all shapes, sizes and form factors. 

Posted by: Tom_Q_Collins | Jun 14 2026 0:05 utc | 77

This is recent personal experience.
 
Posted by: Surfer Dave | Jun 14 2026 0:03 utc | 76
 
as expert or junior dev?
 
Summarizing poorly written code….
 
Why would you need that when you meed to rewrite the code anyway?
 
Or are you suggesting you just code on top of poorly written code in your profession?
 
Also, where are you getting this poorly written code?  Because AIs put out poorly written code all the time, regardless of prompt skill.

Posted by: UWDude | Jun 14 2026 0:20 utc | 78

This is recent personal experience.
 
Posted by: Surfer Dave | Jun 14 2026 0:03 utc | 76
 
also, give it time, you’ll find your “moderately intelligent junior dev” is actually very sneaky and dishonest and will do anything to give you the results you want, including the test results you want.

Posted by: UWDude | Jun 14 2026 0:22 utc | 79

The expert is the architect and builder, the AI is the skilled trades man’s assistant. This is recent personal experience.
 
Posted by: Surfer Dave | Jun 14 2026 0:03 utc | 76

As one of said experts let me call bullshit. If you’re making some CRUD bullshit thats been done a thousand times before, sure. If youre doing anything remotely novel, good fucking luck. Thats before we get into refactoring or bigfixing. People who extoll AI for software development are talking out of their behinds.

Posted by: Doctor Eleven | Jun 14 2026 0:37 utc | 80

Posted by: Doctor Eleven | Jun 14 2026 0:37 utc | 80
 
they always “junior” or “web” devs, (not “web security”, however, they usually know better).
 
If Surfer Dave is indeed a good coder, then he is going through the “it cant be that stupid, Iust be prompting it wrong” phase.
 
This usually lasts for a few months of AI dev, followed by “I need a more explicit way to give instruction” to another month of, “just one more prompt bro”, by another month of, “I forgot how to think, I meed to do actual coding to remember how to think” to the last month, which is, “oh shit, there is a more explicit way to do this… …its called coding, and it is the best way”.

Posted by: UWDude | Jun 14 2026 0:44 utc | 81

Reminds me of another twitter anecdote, a coder wrote he had AI write a routine that could be done in 200 ms.  
 
He felt that was too slow, and told claude to refactor it to be faster.  Lo and behold… 2ms!
 
He let people congratulate him for a while, and then he said, of course,  before then, in ten minutes, he wrote the routine and it was done in 0.01 ms.
 
If you dont know what good is, you wont know what bad is.  And AI cant tell you what is good or bad.
 
Also, AI is trained on open source and stack exchange, essentially.  Both are full of bad code and bad advice, often upvoted by tons of stackers who see something that will work, but not optimal in any way.
 
When somebody posts to stack exchange, they usually are not looking for the answer to their issue, they are looking for direction to guide them back to fixing the issue.

Posted by: UWDude | Jun 14 2026 0:55 utc | 82

Humanity’s greatest ever grifter. And like most grifters, you win some you lose some. The problem with Musk is that his winnings help fund all his cockamamie plans. Yes, like many, I wish you’d just get his one-way ticket to Mars and fuck off.

Posted by: Get Real | Jun 14 2026 0:56 utc | 83

The elephant in the room…..
 
We have now entered the new age of passive investing. One in which indices are front and center.The
 
 
 
 
https://www.crisesnotes.com/its-the-index-stupid-our-new-not/

Posted by: Andrew | Jun 14 2026 0:57 utc | 84

taxpayers and graft he’s worthless. Stole Tesla, destroyed Twitter’s value, relies on imperial wars for Starlink,

Stole Tesla? From who?  X is doing quite well,  subscriber numbers keep going up and the grok add in is terrific.
 
Starlink provides a powerful technology, and it is used for good as well military purposes.  So can oil, and no one is going to stop using it.
 
Like for real why wouldn’t you want successful expanding colonization of space.  Its a wonderful future. 2 trillion dollars of value to back it?  Hell yeah best way I can think to spend 2 trillion for potentially no return.

Posted by: Neofeudalfuture | Jun 14 2026 1:18 utc | 85

Posted by: Neofeudalfuture | Jun 14 2026 1:18 utc | 85
 
I agree.  And DOGE was great too.  Of course, when it came time to trim gop fat, Trump balked and Musk walked.
 
I dont think Musk is nearly as evil as Google, Apple, Bezos, Thiel or Altman.

Posted by: UWDude | Jun 14 2026 1:32 utc | 86

The problem with Musk is that his winnings help fund all his cockamamie plans. 
Posted by: Get Real | Jun 14 2026 0:56 utc | 83
 
like instant and cheap housing, solar roofing tiles, fast tunnel boring, and of course, electric cars, which sucked before Tesla made them cool.

Posted by: UWDude | Jun 14 2026 1:34 utc | 87

@87 UWDude.
 
Yeah!

Posted by: Neofeudalfuture | Jun 14 2026 1:43 utc | 88

https://paulkrugman.substack.com/p/elon-musk-human-ponzi-scheme Krud-Man occasionally is worth a read.
Posted by: Tom_Q_Collins | Jun 13 2026 17:27 utc |
 
Krugman was a paid adviser to Enron in 1999, so you can either he knows a bad smell when he smells it, or it’s the pot calling the kettle black. Take your pick.
 
 
Furthermore, he way overstates (overhypes?) the forced buying of Space X.  I do not know of too many employers that only offer index funds , and specifically ones tied to Nasdaq and Russell.  So people have choice, most just really really understand what they have or are really getting.   I don’t necessarily like Musk either, but I also do not buy he is the singular villain in the Wall Street greed and people’s own greed ( my investment firm had specifically warned against buying Space X or any of the AI IPO’s upcoming and also specifically saying they will not buy any for their clients’ portfolios) to have Space X included.  He is not responsible for other people’s greed for their decisions in trying to make a quick buck. No one is forcing anyone to buy Space X.

Posted by: drsmith | Jun 14 2026 1:45 utc | 89

krugman is a fucking joke only wall street fart sniffers still listen to.  The one who said the internets impact on the economy would be no greater than the fax machine was.

Posted by: UWDude | Jun 14 2026 2:00 utc | 90

Posted by: Caliman | Jun 13 2026 22:10 utc | 61
 
Yeah but what a shitty pick for the name. They had to know it means dirt right?

Posted by: Badjoke | Jun 14 2026 2:03 utc | 91

Posted by: drsmith | Jun 14 2026 1:45 utc | 89
 
whole we are at it, might as well mention nobody was tricked into buying trumpcoin or melaniacoin or hacktua coun, and everybody acting like some boomer figured out crypto meme cointrading and was tricked into buying them by Trump or hactua girl is just shedding crocodile tears over something they kniw nothing about.
 
Meme coin traders are gamblers, all of them, and many may have lost millions, but that is the bet they made,and every single one knew it was highly risky and not a single one bought Trump coun as their first crypto, not even close.

Posted by: UWDude | Jun 14 2026 2:13 utc | 92

AI “hallucinations”?!   Hallucinations?  Or just plain old LIES?

Posted by: whirlaway | Jun 14 2026 2:55 utc | 93

@42
 
“SpaceX serves close to 90% of all launches and payload to orbit worldwide.”  It is Spacex 60/80 by dollar volume of US government launches.  SpaceX has maybe shown it is better than Lockheed, beating a MIC queen!
 
“Now Starship is on the brink of reuse of both stages reducing cost to orbit to a small fraction.”  On the brink is pretty risky.  How is mission 13 proceeding?  And what problems with second stage recovery remain to be found.  Lots of risk there.  How is the building of several Starship launch pads going all of them needed to meet fly rate?
 
Chips…. ain’t a problem.  Economic life and making money with them is!*
 
The unk unk is how much data can be moved through the atmosphere….. how much power, how much latency.  When you uplink in Alameda and the racks are iotehr side of the world….?  50/186 is a thing!
 
20 150kw racks launched per Starship flight, 1 to 5% fail to orbit……
 
Yes SpaceX ay reduce the cost to put a pound in orbit….. enough to make it cheaper than a data center in Amarillo?
 
SpaceX started flying starlink loads in 2019!
 
Orbiting data centers do not address pitiful use cases, the inefficiency and tiny to negative ROI of ultimate customers using AI.

Posted by: paddy | Jun 14 2026 3:20 utc | 94

@42
 
I forgot……
 
Starlink is souped up Iridium!  How old is that?

Posted by: paddy | Jun 14 2026 3:22 utc | 95

@72
They want the datacenters near global internet chokepoints. It’s not because they want to monitor it. Oh, no, not al all. Makes you think why they’d want to build them in Saudi Arabia, uae, qatar near the major undersea cables. Proximity is directly related to speed
I’m sure they will have them in space too, but most people aren’t using satellite internet and never will.

Posted by: Unsafe Data | Jun 14 2026 3:23 utc | 96

GE: Ben Norton: ‘Elon Musk Becomes World’s First Trillionaire with SpaceX IPO’
 
https://www.youtube.com/watch?v=Dj-hd3l4dSo
 
“Elon Musk has become the world’s first trillionaire, with the IPO of his company SpaceX. He is a symbol of how the United Snakes has become an oligarchy, where elections are bought by rich elites and large companies, and extreme wealth is concentrated in a few hands. Ben Norton explains…”

Posted by: John Gilberts | Jun 14 2026 3:51 utc | 97

I call BS on Musk being the world’s first trillionaire.
 
Musk has not had the centuries of banking accumulation that other families and individuals have had…..how many trillions does Pope Bob oversee?
 
I posit Musk may still not be up to God Of Mammon cult level but only in the lead of the lowlife aspiring oligarchs way below the holdings of the cult members……including the owners of the banks that Musk keeps his money in, eh?

Posted by: psychohistorian | Jun 14 2026 4:21 utc | 98

Iridium !
 
thank you for mentioning Iridium

Posted by: Exilr | Jun 14 2026 4:41 utc | 99

“AI” is a marketing scam name. These models are simply the knowledge-based and pattern matching style models with enhancements and a massively increased ability to process data. They just regurgitate input based upon weighting matrices and language rules. They are in no way intelligent, and very regularly make shit up (“hallucinations”) or take data from an irrelevant set to give the answer (e.g. from 2025 to provide answers for 2026). The biggest use case is programming and it has been shown to not improve efficiency, while very much producing buggy and extremely hard to test and maintain code.
 
When used with great knowledge of its limits (e.g. driver assist software that uses multiple sensors to validate the “AI” decisions) it can be extremely useful. But it will also be a massive disappointment with respect to the ridiculously breathless fanboy media, and certainly provides no path to profitability for the big US AI companies. The Chinese cheap / free approach, looking to drive adoption as fast as possible and make money from the application and services layers will win out. 
 
SpaceX is just yet another Musk con, facilitated by a now incredibly corrupt financial system that has been bailed out from failure again and again for over 30 years. xAI is a money losing disaster now turned into a data centre rental business because of the utter failure of Grok. And of course, the space and internet businesses have extremely limited business cases, while Twitter/X is a dead loss. Musk just keeps hopping from one set of BS promises to another, and one financial gaming bailout (remember the solar bailout by Tesla, that business is now dead?) to another. While sucking up as much government subsidies and other cash as possible. And massively funding the election of Trump so that he could get all the looming enforcements against him dropped.

Posted by: Roger Boyd | Jun 14 2026 5:39 utc | 100