Companies usually don't spend that kind of money unless they know there is a return on investment from something.

You're misunderstanding my point. I'm saying that the business and investment community AT LARGE is convinced that AI will be huge, and soon. That is the only reason for the spending.

This is the delusion of crowds. "EVERYONE SAYS" has become certainty. It's massive social proof.

It doesn't make it fact.

I'm claiming that this certainty of AI return WILL collapse on price discovery which means quarters of revenue growth or not. There's now so much money, so much reputation and so many portfolios at stake now that nobody dare question its reality.

These data centers will eventually be repurposed as collateral from all of the failed AI startups. They make the most sense as government surveillance data processors.

Companies usually don't spend that kind of money unless they know there is a return on investment from something.

You're just supposed to put it on your resume.

Its hard for me to believe that companies have invested billions of dollars in data centers and LLM AI based on a theory that it evolves into AGI. I'm not discounting LLM AI's abilities, but big companies are in a frenzy and taking huge risks on this technology. They must know something about it that we don't.

It's not a theory, it's a commonly and widely accepted logical fallacy.

I'm an engineering guy. I know that LLMs basically produce maybe useful results when they feel like it. And I can extend that understanding from one LLM running on my local computer to hundreds of data centers. Most C suite executives didn't come up through engineering. Therefore they just don't know that it's really not real AGI.

I think across the investment economy most players, investors included, are under this fallacy.

I think the end game is this:

- Eventually the AI investment ecosystem will collapse through discovery (seeing that no real path to revenue growth exists.)

- AI startups will go bankrupt and the part of the stock market related to AI will dive south.

- AI data centers will be abandoned.

- LLMs are very useful for sifting data, correlating patterns, etc. Most data centers now being built will be repurposed for government surveillance purposes.

I think the AI bubble is basically creating infrastructure for surveillance, but that will happen after the investment boom is over and those assets sell for pennies on the dollar.

I had it set to strictest permissions, ask for everything.

It kept trying to invoke node.js. I don't have node.js installed on my PC. It burned through 1/3 of my free quota trying to invoke node.js.

It burned the rest of my quota trying to invoke Python. I do have Python on my PC and told it the path. It kept failing.

My task was getting an implementation of text-to-image using an API call using Google's library. It made a whole bunch of mistakes.

1. The consumer and enterprise API have different features. It was giving me enterprise code but I have consumer.

2. Google deprecates and removes its APIs every few months. It was coding to a deprecated API.

3. Google gives you zero images on free tier, so it never would have worked anyway even if I got finished.

I uninstalled antigravity. Maybe I'll give it a try again in a year or two.

The editor was buggy. The installer said "you should install as user not administrator". Then when I ran it, it said "cannot auto-update, must be installed as administrator".

On the GUI, you could change the permission levels, but it didn't remember the choice after you closed the dialog.

Such basic errors made it look like nobody ever tried seriously using it.

One thought that I had is Elon Musk could somehow use his Neuralink technology and a human to act as a bridge to AGI. I can't help but think back to that 1992 movie "Lawnmower Man". Back then it was a horror movie, now we could be living it. I hope something like that never comes to be, but I wouldn't be surprised if they're trying.

How do you get from LLM AI to AGI? Will computing power do that? Is that why companies are building massive data centers everywhere? It can't be to just sustain LLM AI.

I believe more like, if you string 1000 pieces of shit together, then life will not spring from it like the primordial soup. You just get a mountain of shit that's connected like a rope.

Thinking that compute power will make LLM scale up to AGI is basically believing in transmogrification. It won't happen.

In fact if LLMs are chained together you're probably going to run into massive entropy of information, so that what comes out is a lot worse than running the LLM in a simpler mode.

From what I've been reading and viewing it seems like all of the decision makers and investors and capitalists who have developed and pushed the current state of the art for AI are counting on exactly what you said.

IE, the idea seems to be - throw so much compute power at large language models and predictive AI that artificial general intelligence eventually emerges. That's the "industry concensus".

It won't happen.

This is the buried problem of the AI bubble in the stock market and the circular investment of Nvidia in the AI companies who buy chips from Nvidia.

What is being reported in real life is that companies using AI across a range of applications from software engineering to accounting basically don't see identifiable improvements in results or higher productivity. They see high levels of human labor to fact and error check the results. There's an extremely thin market for AI technology and few companies see benefits.

Also consider also that SpaceX's IPO is wrapped around xAI, an AI specific division, as well as plans to deploy orbital data centers. Almost all of SpaceX's IPO revenue is speculating on AI becoming transformative to every part of the economy.

I think:

1) Hardware technology not yet known must be developed that supports AGI. It's probably not just a software problem. (Personally I believe that the real solution that results in AGI will be some kind organic technology, like cultured brain cells that are interfaced to systems.)

2) Almost all of what is predicted about LLM based AI right now is fraudulent pie in the sky.

3) We're probably headed for a monumental market crash. Since most investment money is tied up in AI.

Because the code is unreliable yet is so massive that humans can't check or audit it.

Feature, not a bug :)

The way it is now, I could steal a mail-in ballot, walk in and vote as that person and the election workers wouldn't even ask me for an ID. Democrats are against voter ID because it takes away their ability to cheat with fake votes. Democrat run states maintain their one-party rule by using unverified mail-in ballots to win elections.

Its so important that this act get passed to stop the fraud in our elections.

LLMs can't solve problems they haven't seen before, even though they sometimes get lucky. LLMs aren't aware of their own limitations, hallucinating an answer instead.

In the 60s, an expert level chess program would have been impressive AI. Now Stockfish is free, runs on ordinary hardware, and is nothing special.

"AI" only refers to problems that aren't fully understood or solved yet. Problems that have been solved usually aren't referred to as AI anymore, such as chess programs.

I've been reading a lot about what current AI is capable of, what it can't do well, and what the path forward is.

This guy has a really good explanation of the state of the art today and what I wrote matches it: https://youtu.be/X_nWKJg_D6Q?si=omTrQtkJExp-rnWn

First of all, what is currently called AI and is "available everywhere AI is sold" is large language models, or LLMs. This includes everything such as Grok, Deepseek, and all of the AI tools that developers supposedly use now.

These all have the major restriction of just being predictive generators of new data. In other words current mainstream AI is simply a robot that creates the most likely Nth piece of data (letter, word, pixel or whatever) by guessing based on the previous (N-1)th and earlier data.

LLM AI inherently actually clumsy and crude. It looks like something approaching reasoning is going on but it's not even following boolean logic. It just mimics whatever data it was trained on.

Because of this current AI is inherently non-deterministic. It can return unpredictable results and it generates output that often rarely matches anything in the real world. It can't create and it can't do anything like bookkeeping reliably where precision or final results matter. AI also has the tendency to hallucinate which means that will create garbage, facetious results that don't model reality.

IE: if you used AI to design a machine, it wouldn't be capable of implementing something that actually works. Something would clash and not be physically buildable.

Better chips and Elon's space based server farms won't fix the problem because scaling up compute only speeds up the generation of hallucinated garbage.

AI today is useful for tasks such as looking up or comparing data, brainstorming alternatives, rewriting text, processing images for artistic effects, etc. where precision and real world fit isn't mandatory, and where a human can audit anything that matters.

What is now coming out is that AI being sold to companies to supplement developers are many times more expensive than human developers. Because the code is unreliable yet is so massive that humans can't check or audit it.

Anyone claiming that this AI will do anything more is ignorant and/or bullshitting and scamming.

The supposed big breakthrough that hasn't happened yet is AGI or Artificial General Intelligence. AGI would reason and create with deterministic real world results.

I see this a little bit like creating life in a test tube. It's not possible yet. AGI doesn't exist. AGI means that the machine can really think and create reliable results in the real world.

AGI would be the literal thinking machine that could write an entire book or design a circuit or a building, or manage people, organizations, companies, or economies.

AGI right now is science fiction. All of these videos that have come out that talk about the rise of the machines and killing off humanity as inconvenient are assuming AGI works.

"Generative AI" just means AI that generates something like text, source code, or images. It uses the LLM concept. It's not the same as AGI.

https://www.zerohedge.com/technology/most-important-ai-experiment-youve-never-heard

This article highlights why AI can never be allowed to run something on its own. As smart as AI is, it doesn't make good decisions, no matter which system you use. The results of this experiment are funny, but they make you wonder if you will ever be able to trust AI with greater tasks given how reckless it can be.

Mamdani has the backing of the DSA that is hell-bent on taking the country over with socialism/communism. I wonder how far he'll be able to go before he gets slapped down by the PTB's?

It's a good plan.

- prevent rent controlled/stabilized apartments from raising rent
- landlord's expenses rise due to inflation, but he can't raise rent
- the only way the landlord can stay afloat is to start cutting corners on repairs
- landlord gets fined for not being in compliance with regulations that cost more than his profit to follow
- Mamdani/city seizes property
- property is given to "responsible" people to run, i.e., people who contributed to Mamdani's campaign

https://x.com/VickieforNYC/status/2059433761167569129