Ironically, I find myself with a grin on my face as I read the recent media reports about how the data processing demand behind AI is beyond the scope of financial sustainability.
For several years I have asserted, accurately, the business model for social media was never feasible because the data processing demand needed for the scale of simultaneous users was beyond the capabilities of the revenue side of the equation. I have been told by all the high-horse experts on the matter how wrong I am. However, each story they write about the prohibitive cost of AI proves I was not wrong.
CTH watches the tokenization and subscription fees for various AI model use with the same perspective CTH viewed over a decade of false claims within the financial market that told lies about social media viability and data processing costs.
Now, we watch the seemingly exponential growth of AI capabilities and associated costs with the same pragmatic perspective.
Robotic pool cleaners were introduced two generations ago. Did the pool cleaner business dry up? No, it expanded. Robotic vacuums broke into the popular household appliance market five years ago, you probably have one, did it eliminate maid services? No, still growing.
AI can now write its own code to generate outputs. Are software developers getting fired? No, demand for software designers and engineers is up 15% in the past year.
The mainframe approach, the one AI brain to run all systems, will never work – it is cost prohibitive (see first paragraph – wash, rinse, repeat). Deny this reality at your own investment risk. If needed, politely absorb the ridicule – for it matters not.
CTH predicts AI will become a localized and optimized sub-set for each sector of the economy, requiring each major organization and corporation to adopt specific cost/benefit data libraries and networks for use and functionality.
At scale, a thousand coders each working on Gemini, ChatGPT, Anthropic, Grok, etc. will become 100,000+ software designers working inside companies to create personalized, targeted, bespoke AI data systems and networks; each system specifically tailored to the industry or sector of business. The intranet of internets will happen again.
Creating and selling AI system networks and integration functions that are personally tailored to highly specific company functions, creates an entirely new sector of the technology industry that has not even begun yet. [There’s an investment opportunity there]
Will AI robots replace some repetitive human functions? Yes, the ice rink Zamboni will likely not have a steering wheel, just an emergency joystick. A reference for a comparative industrial scale Roomba vacuum, or the robotic pool cleaners. However, at scale the robotic industry is slower than human efficiency in almost all sectors that matter; the cost benefit analysis will limit growth. The maid service sector will not be impacted any more than the software developers (see chart above).
It is not an issue to fear some AI task efficiencies will grant more time available that will be filled with alternate task capabilities. Human productivity will increase in certain sectors of the economy, but humans will not lose work opportunities. Blue collar jobs will continue to expand as each of the hardware tools developed will need manufacturing, installation, maintenance and monitoring.
The further downstream the worker is from a repetitive function within the [XXXX] industry, the more irreplaceable they become; remember that.
As to the bigger picture of fully developed AI and the intersection of information and knowledge; yes, the automation of AI can present an issue. However, all AI concerns can be mitigated so long as multiple, alternative AI systems exist within the larger information realm.
As a nation we need dozens of different AI models each competing within the industry for the best AI product. As long as we have multiple AI systems, alternatives to the hive-mind, we do not need to fear the AI network as a source of information. If we don’t like the AI outputs, we can switch to an alternate AI provider.
If the subscription cost of the AI is too high, then as long as we have a competitive market where a lesser expensive, perhaps bespoke, AI option can exist, we should be okay. Let the free-and-fair market decide.
If AI outputs don’t offer empirical truth or real value to the end user, we should be fine as long as consumers have alternative options available. AI providers should be information providers in the same concept as cell phone providers. The key is to have multiple, competing AI systems available for industrial, business, professional and personal use.
On the upside of this information worry dynamic -in the pragmatic and optimistic perspective- we have the cost limiting nature of a massive singular AI information network.
A single AI central brain handling over 360 million users at once, all requiring identical responses that update with every tiny change in a multi-trillion datapoint-per-millisecond data stream, is far beyond the capacity of any computational AI system. The costs tied to such a setup are only now becoming clear, and AI business models are starting to fall apart in real time. This is a hard truth that isn’t going to change.
Within the AI business, those who can carefully write AI input instructions to achieve maximum value in AI output -industry by industry- will become increasingly more valuable. Those who can train AI to be cost effective -and provide materially beneficial outputs- within their granular sector of business, within each company, will become priceless to the organization. Wage rates will follow competency.
As noted by David Sacks in this segment highlighted below, the one key about AI to emphasize is the need for multiple competing models. If China (hive mind) has their model, and Europe (another hive mind) has their model, and the United States (entrepreneurial competitiveness) has multiple competitive models – we will win and simultaneously we will retain freedom.
What we don’t want is a singular AI model to win the support of the United States government and then end up with an AI regulatory system where they start defining terms of “safety” to eliminate information adverse to the interests of the government that regulates it. Both China and Europe will predictably do that.
Sundance, I will NEVER forget your prescient and informative thoughts on the backroom overhead enabling online platform operations.
This post, like many, arrives like a refreshing wind on the Sahara. THANK YOU! 🙂
Stop the bus I want to get off.
“AI” is mostly a hoax. Much like quantum computers.
There are LLMs, which are essentially just nested “near” regular expressions. They have no reasoning ability at all. Any resemblance to reason is just the logic built into their source material.
“Reasoning” models are basically the same LISP style decision trees developed back in the 70s with an LLM running on top of it.
All of these are useless for any significant business application.
YCombinator is a good source for info on this since everyone is a software engineer. None of us are particularly impressed with AI.
I am not aware of a single business process that has been solved by “AI”. Most “AI” tools that are doing absolutely anything are just plain hard coded automation. I’ve been doing that for 20+ years. A spell checker is not AI but if it was a new feature in 2026 the company would call it AI.
Climate hoax to AI hoax. I have already seen mention of AI credit systems.
How about military? Targeting?
“What we don’t want is a singular AI model to win the support of the United States government and then end up with an AI regulatory system where they start defining terms of “safety” to eliminate information adverse to the interests of the government that regulates it. Both China and Europe will predictably do that.”
If the Democrat-Communist Party ever steals another National Election and the Executive Branch, we will immediately and ultimately become pawns of their Chy-na paymasters (God forbid).
Jack Gamble from Nobody Special Finance has been screaming for years that the AI bubble is largely a giant circular accounting scam. According to him, Nvidia, CoreWeave, et al are going to go down hard.
In my area, they’re talking about building data centers in King of Prussia and Conshohocken (business friendly Montco communities just outside the city). Nobody wants them. Josh Shapiro, our gov who wants the Dem nomination for POTUS, has come out in favor of them. We’ll see what happens.
It is. NVidia “invests” 100b in CoreWeave, CoreWeave commits to buy 100b of Nvidia cards. NVida’s valuation goes up 100b, CoreWeave valuation goes up 100b.
This has happened about ten to fifteen times with several of them, especially NVidia and ChatGPT.
And when the crash happens, you will get to bail them out.
Modular versus centralized.
Individual liberty versus Globalization(big govt)
Pulling turnips because the govt says so or self actualization.
Same old battle, same foes. Rise and rise again until lambs become lions.
What is starting to be of concern is another economic bubble centered around AI and data centers……thoughts Sundance?
Great post, Sundance, as always.
“all AI concerns can be mitigated so long as multiple, alternative AI systems exist within the larger information realm.”
Let’s pray that is the case.
(I’m afraid I can’t let you do that.)
Fifth paragraph:
<Robotic pool cleaners were introduced two generations ago. Did the pool cleaner business dry up?>
🙂 🙂 🙂 🙂
Wet dream ?
AI=Solyndra.
This is little different from the nuclear power projection excess in the 1970’s that is about to be repeated today. The cost will be grossly underestimated, the demand grossly exaggerated, and plant after plant cancelled. Don’t worry about data centers too greatly, few under discussion will be built and operated.
I’m an old school engineer and software developer who has adopted AI tools as mandated by corporate. First it was Copilot, now it is Anthropic/Claude. My token usage this month of May was $0.93. That’s right, and it was worth it! I asked Claude very specific things that I didn’t want to do but could do, and it did it. Such savings! So impress! Will I have a job for a few more years? Gotta ask a question: How many C and C++ programmers, at a system engineer level with background in electrical engineering and advanced mathematics are there? Today? Do you want to pay him or (unlikely) her a decent low six figure salary, or do you want to bank on Claude to keep your 160+ Linux servers running custom company software?
Yet to see Claude or Qwen manage to fix a bug or integrate a new feature into existing code sets. Doubt it is forthcoming.
The problem with code has always been design and not planning for or correctly dealing with potentially conflicting inputs. Claude is atrocious with this. Any update would create more problems than it would fix.
The biggest hurdle I see in Ai Sundance is our infrastructure.
The idiots like Al Gore and Greta have made it so we didn’t build electrical infrastructure.
We shut down coal plants, didn’t expand on Nuclear tech and acted like the freak doomsayers so we didn’t expand our energy needs.
All of this stuff should’ve been in place and expanded on years ago.
The greeny weenies hurt our economy.
We have a lot of catching up to China in the energy sector.
I ask people to not be afraid of Ai or data centers because these things are coming anyways.
We might as well accept it and use it to our advantage like those who were the tech greats of the past.
This reminds me of the video industry screaming the VCR would kill Hollywood. It did not. It made the total market increase exponentially.
Same thing with CDs. And with home theater systems.
Basically, it’s fear of the unknown and companies talking their book.
Hollywood is dead now. Who goes to the movies anymore?
I travel some international and I think AI has improved the experience.
I have to enter my flight itinerary on a government app before I leave. Instead of waiting in long lines for immigration and customs I breeze right through. My photo gets taken and I’m done.
People will squawk about privacy but the old days of handing a simple ID are done. We need to know who the person really is and that requires biometrics.
Classical Greek sources describe a working steam turbine, and descriptions of a steam powered device appear as early as 20-30 BC. Although humans were experimenting with steam power at least three decades before the birth of Christ, it would be another 1,800 years before the steam engine ushered in the Industrial Revolution.
When I look at the developing AI industry, particularly in regards to its prohibitive- perhaps currently unsustainable- costs and the small but growing skepticism and resistance to it, I think about those 1,800 years. That is not to say that it will take 18 centuries for AI to revolutionize human civilization, but the Greek turbine is a reminder that technological development is not linear, that its potential is not always guaranteed to lead to adoption by and integration into society, and that the very conditions which create technological innovation can sometimes prevent those innovations from having practical impacts.
I use various AI platforms daily and it’s the most amazing new tool I’ve had in my 55 years. That said it has a long way to go. It’s in its infancy. But it’s getting there, rapidly.
Just as the Articles of Confederation and the Republic was scrapped, so too will competing AI models. Social media wasn’t economically feasible until ad revenue and the sale of user data.