"AI" is it not the centralization of the internet?

While using "AI" tool principally Gemini on android phone when I need help with mathematic, it basically replaces my teacher which is available by chat but not 24/7, I kind a have the feeling that it's like the social media that centralized the attention of the world in their website but this time the feeling is that the whole internet is available information is centralized in the Gemini app. Isn't dangerous to accustom ourselves with it ? What do you think about this new technology, I'm interested to have your opinion?
 
It feels like we have a powerful technology break through, or call it a major technological evolution in the way of processing data or computation, but the social context in which it has appeared make it dangerous. Or is it the same circle that renew itself ? Should it be not allowed that this tech be possessed by individual and for a capitalist purpose considering that it feed on the entire history of the human species is collective data ?
 
when I need help with mathematic
TI-83.png
 
Recently CHATGPT recommended to someone that they should stop all of their daily eating of "salt" and instead eat "sodium bromine". So he bought "sodium bromine" over the Internet and ate that instead of salt. Now he has "Bromism", and although the article reads like he might survive it, it probably isn't fun. Article here: Don't eat Salt, eat Sodium Bromine.

And this was (from the article): A 60-year-old man who had a "history of studying nutrition in college"

The usual response we hear on this is "Well, that was stupid thing for him to do !".

What are your thoughts on this?
 
What are your thoughts on this?
"Well, that was stupid thing for him to do !"
That guy his actions, I didn't read the article, in my view are representative of the spirit of consumerism. He just consumed the information provided to him, blindly trusted it, I would be curious to know if it's a premium user or a standard user, and if it's a MAGA retard or not. For my part, I refuse to pay anything to use it, and I never thought it's worth the price anyway.
 
But this topic is not intended to be a troll, but to have a philosophical and a fact based opinion about this technology, and it's impact on society and the politics. I don't want to know if you hate this tech or not if you can't explain it pass your first emotional state or biased state of mind. It's real questions that I ask here.
 
It's been discussed before here. I consider AI, as used in your example, to be a search engine. It helps gather facts and non-facts from sources already on the internet into one digestible document. Just like any internet search, these facts must be researched further and not taken on face value. Sometimes they are good enough if that's OK for you.
 
It's been discussed before here
Not the way I put it. And anyway it can be discussed again ! I agree not to blindly let your self be lobotomized by this tech. It is written just at the downside of every chat input and output of the "AI" tool. I use it only for mathematical concept and I ask a lot of precision, so the chat is normally pretty much interesting, I don't use it for research on important things or for argumentation because in that domain the AI tools is pretty much locked and boring. I know a lot of people in this forum have a bad opinion about this tech which is pretty much mainstream and due to the companies that hold it., but I want to discuss it more profoundly. The opinion that everyone seem to easily have is limited to the experience that we are allowed to have with it, so the social context, in my opinion. To reject totally or to accept it totally is the same way of being biased. I let you all sit on this topic and if you are able of reflection, to have a thought about it, than come back and share it !
 
I’d avoid AI especially ChartGPT-like for serious tech knowledge learning tbh. I remember the days I was trying with ChartGPT to help me learn eBPF and it automatically generates tons of code looking pretty good to me, then I find out a lot of eBPF syscalls are just made up and do not exist.

I mean daily casual questions are totally worth it, like how to cook a potato. But I would not risk again trusting ChartGPT to help me on any tech related issues. The best case, it doesn’t work. The worst case, it crash the Prod.
 
I think A"I" is just the latest fad. Sadly, I think this fad will stick around because there are enough gullable people to continue falling for the same scam over and over and over. SOME of these tools have a use, many are useless, many are harmful (Recall), and this entire AI thing is overall just the latest techbro craze.
 
I'm sure everyone notices every day how many articles and ads now hype AI usage in our future and how products now use it.
And I hope everyone checks the results it produces.

Since duckduckgo.com offers to use some ChatGPT for free I tried it several times to find something. I summarized my experiences under "It's all about jokes, funny pics..."
This sh... really simply just makes things up, not checking them itself, but apologizing a lot if you point out that it was garbage it told you, and then tries to distract you to complete other things, like shopping websites.
It's like talking to a pupil you caught on not did his homework, but believes he was smarter than you, and he could score by telling you other things you didn't asked but he knows. 🤪

I think A"I" is just the latest fad.
I wouldn't put it this way.

AI simply is a technology that recently reached a new level of its development.
When a technology development reaches a level even Johnny Everyone can do something with it - a useless toy was even better than doing something useful with it (in our society. which leads to social questions like decadence, et al); CShell brought the example of the nutrition educated who ate poison just because a computer told him so. When in the late 1990s the first (mobile, PDA) navigation systems for cars came up I experienced several people got lost on the exact same way they drove daily for twenty years - then money comes in, and with it come business people. There is some shit that can be sold large.
So a hype is produced. Euphoria is needed. Because shit needs to be sold quickly before the majority realizes it again ain't the jack-of-all-trades wonder solving all problems for free, and bringing a lot of money for everybody without effort. Even people calling themselves nature scientifically educated laughing about the stupid people from former ages still want to believe in magic.
Hype means mixing up things: hopes, wishes, dreams, and fancy ideas with the realistic, actually useful but boring stuff, like it's to be used by experts with expertise simply as an assistance within very specific, and limited applications, not really a replacement for humans in general.
Herein lie the real problems hypes produce: They produce imprudently rash misses, the need to later clean that up again, to separate and dissolve all the muddle, while at the same time there is a lack of scientific ways to talk to each other, but avoidance of putting things into perspective.

But by what I understand from the news I see on HN, et al, to me it's clear:
This bubble already has begun to burst.

Sad is only:
People again will not learn nothing from it.
It's just a question of time the next bubble will be hyped, and (almost) everybody will become euphoric again about another, new technology that promises to make humans expendable. :-/
 
There are already stories of companies hiring back workers that they let go, figuring AI would replace them. One made me laugh, they found their client base hated the service giving by AI vs. human customer service people.
That's kind of an obvious one, IMHO.
 
after a chat ai forgets evrything.
Only if you clear your browser cache.

As for info research, if you just use a public service, it's not that different from asking a random person on the street, it's gonna be a hit and miss.

That doesn't excuse anyone from doing their own thinking about the veracity of the information they receive. 👿

AI is already being deployed to con people out of money, provide excuses to not hire people for stupid reasons, and to pretend somebody did or say something (when in fact they did not). As a famous example, If you know how to work the software, it takes about 5 minutes to produce a deepfake video of Marco Rubio making a ridiculous claim of anything under the sun.

Yeah, you can tell an AI to shut up. You can and storm out of the room, and throw your phone in disgust - and an AI won't get pissed at you for that. It will just keep feeding you info and not give a rat's ass about how you feel about it.

You can ask an AI to teach you stuff - but it's ultimately your own brain that determines whether or not you get anything out of that. Whether or not that effort is acknowledged by another human being, that is a very separate matter. I mean, what are you gonna do once you have the skill? It ultimately doesn't matter if you used an AI to learn it or a human helper.

After all, what else can you expect out a pile of plastic and heavy metal that only serves as a mask to hide the devil in the details?
 
A"I" sort off solves one previous buzzword - BIG DATA problem. And the artificial problem called bureaucracy. Net positive is therefore very little. Some say, given the huge investments in data centers etc. it will be a major flop.

I use it basically instead of Google Search, which became lobotomized over the years (takes just a few main words from my query and shows me some generic Reddit posts, ignores the nuances in the question...and of course shows the sponsored stuff first).
I also use it to summarize texts.
Current corporate pipeline:
Concise message -> ChatGPT generates several paragraphs of fluff -> send -> recipient puts it in to ChatGPT -> generate me concise version of this sh...

Similar to learning - it can take a lot of text and rewrite it to your current needs. Not bad. I like it, I use it. But is it really worth the trillions invested?

re: Centralization of Internet
I am a fan of Heinlein. I was surprised he did not mentioned Internet that often, even in his latest novels, when Internet was already invented. Miscalculation? Unlikely, that guy was a visionary genius. He was right.
Internet will end up how it started - Public library. You just ask it for the news and Encyclopedia questions. Social media? Nah, that is for teenage girls, marketers and politics nerds, really.
 
If one is putting up a TI calculator against AI, AI wins.
Put up an old HP doing RPN and HP wins.

Honestly is AI at this point in time, really more than Google plus DuckDuckGo plus Bing?
Can it matter in some spaces? Maybe.
But can AI root cause data flow across complex systems (multiple containers)? Maybe sometimes, but I think you need a human that understands the space to teach the AI.
AI as a tool to help? Sure, we all know the value of StackOverflow, so if AI can make my lookups easier, that's good.
But since AI can't be fired from $JOB (and I can) I still need to vet the responses of AI.

Think of Law: use AI to generate an argument for defense of capitol offense, AI as the judge, AI as the prosecutor, AI as the jury, but a human is on trial.
How can you guarantee the the HUMAN will receive a fair trial when everything else is AI?
 
...but this time the feeling is that the whole internet is available information is centralized in the Gemini app.
Is mathematics taught in universities today through the prism of AI slag?
In my opinion, Perelman proved the Poincaré theorem (hypothesis) without this AI of yours.
True, Grigori Perelman went a little crazy from the overload, but that's another matter.
Charlie, have you thought about just buying at least paper books on mathematics?
It's not expensive, it's accessible and it's always with you, even if the planet's electricity is cut off.
 
Charlie, have you thought about just buying at least paper books on mathematics?
It's not expensive, it's accessible and it's always with you, even if the planet's electricity is cut off.
I Totally agree with that, problem is that I'm broke right now, story of my life. And I know that I can find every book I want online but reading on the computer is not the same as reading on paper. Also don't have much time to read.
 
Charlie, have you thought about just buying at least paper books on mathematics?
After I graduated from college, I hunted down a math book that I spent $68 USD on when I needed to buy it for class. Back then, I sold the book back to the bookstore (as was the norm), but after graduating, I regretted that. So, I hunted it down on Amazon, and had to pay over $120 USD for the exact edition. But now, 20 years down the road, I checked back on the situation. And I was flabbergasted! 😲

When I was in college, I used the second edition of Susanna Epp's "Discrete Mathematics with Applications", hardback. It's $104 USD on Amazon right now. And nowadays, my college is using the 5th edition of the exact same book! That 5th edition is $181 on Amazon... also hardback.
 
You wrote everything correctly. Life is becoming more expensive everywhere on the planet.
It is not difficult to check even with ordinary YouTube bloggers, without any speculative
calculations from the US Treasury or the Federal Reserve.
How much did a hamburger cost in the US in the 60s? And how much today...yes,
we understand everything. :)
There is an excellent millionaire blogger "America Inside Out" (Alex Brezhnev, a Russian Jew from Crimea), he covers your economy well with swearing, because he himself spent his whole life in an iron booth, and for 1 hour of work (in terms of
life) he becomes poorer and poorer every year.
In the 2000s, he earned so much per hour that he could buy, for example, 4 cups of coffee,
and now only 2.
But books, unlike coffee, are with you for life.
 
Back
Top