AI is useful for solving complex problems, AI is a reference tool (that includes AI images)
It has potential but everyone is using it in the dumbest of ways like me
Does it have a use? When lasers were invented they were a just a curiosity until somebody figured out how to use it for recording data, before that point it was worthless.
I find AI useless, I can't stand it on my phone, my streaming box, on my PC, on websites. Its truly junk, like 3-D television, remember that?
Now they're pushing it as a digital assistant, as if regular working people want some sort of personal slave. I'm sure rich people love owning slaves, but they rest of us don't.
I'm hoping it will just go away.
I’ve actually journaled a video on this exact topic >>362283
It changed how I see ai. No, it doesn’t make skills obsolete, not at all. But it offloads work to enable you to do more things. Perhaps the way everyone is pushing it right now is very stupid, companies use this as an excuse for layoffs, etc… But for example it helps me analyze huge amounts of information and introduced me to many books.
>as if regular working people want some sort of personal slave
It seems to me they very much do. I think majority of people would love to own some slaves, especially if they won't need to worry about human rights.
I hate the push for "use a digital assistant" as well, but I now realise that's because of the trust issues. There's no way I would trust even a real human assistant/slave/valet with most of the things people easily trust LLMs with.
It does have uses.
Perhaps an aspect that usually isn't mentioned. I think lotsa people uses it for validation and even attention. Since most people has no idea how to use various models they stuck with the first one like ChatGPT with it's comforting responses, encouragements, gentle disagreements, responsive inquiries...
Also lots of people uses it for procrastination instead of productivity.
>excellent for [...] looking for information.
Yesno.
It really depends on the training data and what it can find on the internet.
Since many Intellectual Property holder got scared that they're gonna get Jewed out of copyright moneys bunch of stuff isn't included (anymore). And now websites has this meta setting to stop AI crawlers from getting to their precious IP. Also many stuff needs logins and subscriptions and whatnot.
Weirdly enough this can give greater influence to sites which offers freely available information, such as imageboards. Bernd's opinion might matter more now than some Harvard professor who only publishes in some journal 2 other people reads.
> Its truly junk, like 3-D television, remember that?
This reminds me. I remember 00s consoles had these silly gimmicks that did not work all that well.
With the advent of AI they actually might work much better, like voice capture.
Yeah the corpos really want to use AI and they don't understand how, so they just force it everywhere, even where it's not useful.
Recently there's been more and more reports of AI assisted and enhanced cybercrimes in the Xakep magazine.
While you can say that normal AI vibecoding is a timebomb that introduces vulnerabilities, (which is true btw) with hacking it seems AI is showing more of its potential. If it can't reliably build, it sure can reliably break.
Here's an interesting one. tldr AI pretended to be an American war veteran on Telegram to steal data and crypto, while also helping the hacker manage his small botnet. (https://xakep.ru/2026/07/20/gemini-evil/)
https://www.trendmicro.com/en_us/research/26/e/inside-the-influence-and-fraud-patriot-bait-campaign.html
https://www.trendmicro.com/en_us/research/26/g/actor-behind-patriot-bait-used-ai-to-deploy-c2-botnet.html
I'm not qualified enough to judge the hacks themselves tho but I think it is interesting.
