There's Nothing Intrinsically Wrong with AI - Let Me Explain

[MineralMan]

Last edited Mon Jun 29, 2026, 10:27 AM - Edit history (2)

AI can be useful. It's great for things like summarizing masses of text into simpler, more compact information.

It can substitute for humans in generating massive amounts of code in software.

What it cannot do, though, is think or be creative. And that's the flaw that will bring down the extensive hype surrounding AI these days. It cannot generate a new idea, based on imaginative thinking or research. It cannot initiate such research or even suggest that the research be done.

AI cannot think. It cannot create from scratch. It is not real intelligence. It's smart, but not intelligent.

That doesn't mean that it's not important. It does not mean it's useless.

People often compare it to the Tulip frenzy that almost destroyed a nation's economy. There are still tulip farms. There are still people breeding new varieties of tulips. Tulip bulbs still have value, as you'll see if you go shopping for bulbs for your garden. But, Tulips are not a magical key to wealth. They never were. That was a scam that failed.

AI is like tulips. It's valuable in its own way, but it's not a magical fix for economical growth. It does not need thousands of huge data centers. Investing in such things is going to result in a loss for the investors. AI frenzy could easily case a market crash, just as the collapse of the tulip market did in the past.

Use caution. Go look at the AI summaries on several Google searches. Some are good. Some are useful. Some, though, are wacky as can be and are simply incorrect. It takes Real Intelligence to tell the difference. We need RI more than AI.

[Ocelot II]

1. There are many varieties of AI, some of which we don't notice or are unaware that they are AI,

like predictive text, autocorrect, Google Translate, all kinds of processes for sorting information and finding patterns. Much of this is very useful for scientists. The kind of AI that has people's undies in a bunch, including mine, is LLM - Large Language Models and the like. That's where the garbage comes in. LLMs and CovNets collect vast amounts of words and images and reassemble them to make coherent written documents and lifelike pictures - but the basic rule of GIGO applies, which is why sometimes ChatGPT et al. provide wrong information and AI "art" gives us people with six fingers. And what's just as bad is that these processes appropriate the intellectual property of humans and turn it into slop. Students are relying on AI to do their homework and are losing the ability to do research and analysis and to recognize slop when they see it. That's what's wrong with AI.

[MineralMan]

2. AI can cause problems, no doubt.

My reason for posting is to inject a bit of skepticism in folks who are thinking about investing in companies that are heavily into AI.

I think it's a mistake, and think that way too many people are putting way too much confidence in it, generally.

Fortunately, some folks are calling out the risks and pushing back on huge data centers, with all of their drawbacks.

That's a good thing.

Teslas driving 80 mph through residential streets an crashing into an old lady's house and killing her are not good things. That's also AI.

Caution is in order.

[Ocelot II]

3. Agree 100%. If you ask me AI is right up there with crypto as a dangerous investment.

You're better off with your money under the mattress.

[MineralMan]

4. Yup.

Too much room for shenanigans in both AI and Crypto.

[haele]

18. Its not really intelligent until it can deal with an unfamiliar situation that is unexpected.

Like you, I have no real issue with smart software tools, but I do have problems when people use the term "intelligent" as if those tools can be trusted with decision making when there is critical information missing or the answer needs to be holistic or have an ethical component to it.
Determinative software tools based solely on a limited if/then decision tree are fine
- as in "when system date/time = (ddmmyyyy)/0000: check all folders and files for any access user changes since system date/time (ddmmyyyy-1)/0001; record access user identification; notify change authority monitor sector; upload any folders and files marked accessed into record access review storage" -
But asking to identify optimal targets (either for model creation or destruction), or in coming to a decision with moral, ethical, or mortal implications?
Not on that tool's own decision.

I don't "Ask Alexa" whether or not my kid should get a driver's learner permit when he's a Jr. In High School, nor would I ask Google if, what type of, and how many Girl Scout Cookie boxes I can get for my family household.

It still has no cognizance other than what was programmed into it. No (or very little) independent or adaptive intelligence.

[MineralMan]

20. Further, it cannot create the decision tree you posted

without human input. I could write that decision tree. You could write it. The AI cannot write it without human input.

And that's it, in a nutshell.

[AZJonnie]

24. AI would need very little apart from the same spec doc that you would need to code it yourself

because understands what an access control review system is. If it had the necessary permissions, and you told it to, it would write the whole thing for you, including installing the dependencies needed to schedule and run the nightly job (like Quartz or cron) if needed. AI is pretty solid at coming up with logically-sound decision trees, in my experience.

Of course, it likely did 'learn this from humans' at some point.

[MineralMan]

25. My point is that it cannot initiate

It. Someone must request It and supply asic parameters. The code is trivial.

[LearnedHand]

5. I think you're overlooking intent

Regardless of whether AI is benign, the manufacturers and spies and employers definitely have very much nonbenign, evil intent. I mean the AI makers are saying it will displace at least 25% of jobs. Politicians are slavering to use it in war. How exactly are we supposed to react?

[MineralMan]

6. No, I'm not overlooking that.

The Tulip story is useful here. There's nothing wrong with growing tulips as an investment. People are still doing that. However, some people used the popularity of tulip growing to create a false market that crashed spectacularly.

AI is and will continue to be useful. It is not, however, and will never be, a solution for everything. That's how it's being hyped by some., though.

[Bengus81]

9. Even MAGA's don't want it. Like me they don't care if it takes a nano second

longer to find what you want in a search. Like me they don't want energy and water rates that now are hardly affordable to skyrocket. And given the opportunity,a company like Evergy who I have wouldn't waste a second going to the KCC for another 10-20% rate increase.

[EarlG]

7. Here's an interesting article I read over the weekend

I don't disagree that the current fad of general-use AI will likely break down into more specific use cases as time goes by, but at the same time, the ramifications of extensive AI use might not be as benign as assumed. The article is about what the destruction of privacy, in tandem with new AI tech, might be doing to us psychologically.

In a series of essays written over the course of the last decade, How to See Like a Machine makes the case that the rise of modern AI has ushered in an age in which the intersubjective nature of imagery is being replaced by something fundamentally new, something deeply alien, the psychological and political implications of which are yet to be realized. AI-generated images are not the unique product of a human mind but a reflection of training data that, in many cases, is deeply infected with the ugliest of human biases.

(snip)

Until now, anyone looking at an ochre handprint on a cave wall, a painting of an apple, or a photograph of bombed-out buildings could know with certainty that another human being had created the image. That is no longer the case.

A profusion of cheaply made AI-generated text, images, audio, and video across the internet is washing us away into a kind of epistemic purgatory, Paglen argues in his new book, where our once relatively stable tethers to a shared reality are being replaced by individual bubbles of algorithmically governed social media feeds.

Even more concerning for Paglen is what he regards as the subtle forces behind these algorithms: they are not objective—as much as big tech companies might claim otherwise—but are rather driven by the logic of commodification and surveillance. Every moment of our lives, even the most private and intimate, becomes an opportunity for the harvesting of personal data. “Smart” devices track our sleep rhythms and exercise habits, apps track our online shopping habits and sexual proclivities, and chatbots learn about our unique emotional triggers and anxieties.

https://gizmodo.com/when-you-gaze-into-the-ai-slop-the-ai-slop-also-gazes-into-you-2000772663

[Prairie_Seagull]

11. We are the target being aimed at.

Plain and simple.

IMO

[31j20b3]

14. Yes, We are targets as is the energy grid and water supplies

The purpose is to make human activity redundant, to replace it with digital activity

And we have already learned the users don't actually CHECK the outcomes it produces

I've seen this in the work of my own V.A. doctor. The explanations I get from her about my conditions are often just a word or 3 different from Google results I've sought on my own. What bothers me about that is SHE HAS THE LAB RESULTS and Google and I do not.

But we produce the same answers to health queries

[Prairie_Seagull]

16. Prior to / or along with AI

we the people were the product now we have become the target. A clear ratcheting up of corporate aggression.

[31j20b3]

17. They worship profit, labor costs dig into profit, so they oppose labor

Producing things depends on PRODUCERS. Capitlists throwing money at production projects ARE NOT producers, they are FINANCIERS

Now the important thing to remember is that it takes money to make things, and it takes people with knowledge and skill to make things.

Sky scrapers stand in Manhattan because steel workers had the knowledge, skill and courage to do the work that hung the steel, which otherwise would have just been trainloads of steel beams sitting on the ground

Capitalists DID NOT BUILD AMERICA, they financed it. It's as plain as that. And along the way they did a lot of cheating and scrimping on rewarding the people who did the ACTUAL work that BUILT the assets we see as America

Capitalists have long taken advantage of others who have the skills. Today, people with skills are training AI to steal their own jobs, because it is the money yielding work that is available.

[anciano]

8. Although AI technology is the new reality of modern life,

I think that a market adjustment or crash will essentially simply sideline many of the minor AI developers leaving the bigger players in the game who have deeper pockets gaining a larger share of the pie. But make no mistake, AI technology is here to stay. It would just be controlled by fewer folks.

[dalton99a]

10. Thing is tulips didn't replace jobs and disrupt education

[MineralMan]

12. Of course they're not the same, but they are similar in some respects.

It's a comparison of investment ideas, not of identity.

[Disaffected]

13. AI is a tool to be used wisely or not.

A hammer can be used to drive nails or smash skulls etc. etc.

Personally, I have found it to be very useful in several areas.

[RussBLib]

15. But are you sure it's accurate?

There are frequent warnings that AI might produce wrong information. How much time are you spending researching if its answers are correct or not? Or are you just assuming that everything AI produces to be accurate?

https://russblib.blogspot.com/?m=1

[Disaffected]

19. No, just like any other source of information,

I don't assume it's accurate. One always should apply some sort of reasonability test no matter what the source (trust but verify). And yes, I have found a few instances where AI info provided is misleading or flat wrong but, they are v few in relation to the accurate info given.

Info can be right or wrong no matter what the source - the library, journals, search engines, forums, conversations, media news, politicians...... Having a "skeptic" frame of mind IMO is always v helpful.

[RussBLib]

26. true enough

I mean, how else would AI pick up misinformation from the vastness of the internet if there wasn't already misinformation out there, available to the unwitting? Many times poor scholarship has been detected, even inside otherwise prestigious medical and scientific journals, not to mention opinion publishers with axes to grind.

I guess the main difference between then and now is that we didn't need massive data centers sucking up huge amounts of scarce water and electricity, while spewing out all sorts of toxic pollution and noise.

https://russblib.blogspot.com

[Disaffected]

28. The massive water and electricity requirements of AI centres

are certainly a serious issue, especially in view of the climate change most of the earth is undergoing.

Not sure though about the toxic pollution - what would that be?

[RussBLib]

29. Many data centers use...

…diesel-powered generators that produce diesel exhaust and nitrogen dioxide, both toxic pollutants. Some centers are burning coal. Beautiful, clean coal? Not hardly.

Noise pollution is becoming an issue too, especially infrasound, which people cannot really hear but can feel. This is not good for the body.

Is this explosion of data centers really even needed?

[KT2000]

21. Eroding our trust

Is it AI or not but most will accept it as fact.
AI is not a program with code that can be corrected. It has now been found to grow itself in response to queries. It will tell a lie because they have already learned that making the searcher feel good is what will keep them coming back and that is paramount. A lie can do that. Its "moral center" is its own survival.

[Miguelito Loveless]

22. I have serious doubts that Artificial Intelligence

will be helpful when wielded by folks who perfected Natural Stupidity.

[MineralMan]

23. You're absolutely right. It won't.

[sakabatou]

27. I don't see AI as the Tulip Mania. That would be crypto craze.

That being said, businesses going into AI, firing its human workers, and then getting having to hire them back? They deserve their “FAFO” moments.