My AI girlfriend has run off with the toaster 
Again
4 Likes
Just using this thread to rant about AI. At work everyone started using it. Some colleagues write every Email, bigger Teams messages or Tickets using AI. Some seemingly tell it to use the maximum amount of emojis possible inside a given text.
AI is the downfall of a normal discourse between humans in the working field.
I can’t be sure anymore my colleagues actually exist…
I might have a different form of AI psychosis…
4 Likes
We have just interviewed newly qualified nurses for Preceptorship positions, they all have to do 6 months (to prove they are not going to kill someone 
), I was told they all use ChatGPT for their applications!!!
1 Like
found the original post of the gemini crashout that was even in the news where it started looping “i am a disgrace” and such things, going completly crazy
https://www.reddit.com/r/GeminiAI/comments/1lxqbxa/i_am_actually_terrified/
1 Like
Haha. That s funny. I used to say the same about social media. What ever happened to face to face dialog. That used to be how it was done…
2 Likes
Eh…yeah the Ai craze just go to show how lazy humans today have become! 
First it was journalists/Media that got lazy not doing any fact checking now lazy people using faulty AI.
We are probably doomed as a species 
3 Likes
Yep. Yet again someone invents a tool. People act like morons using it. Everyone blames the tool.
5 Likes
Which, in my opinion, is wrong to call (artificial) “Intelligence.” They’re algorithms, no more, no less. They can provide results that make sense to us, they can be useful and faster at extrapolating data, and so on.
But they’re not intelligent. They don’t understand what they’re doing, they’re not aware of what they’re doing; essentially, they just perform a task. It would be better, in my opinion, to replace the term AI with algorithms; it would be more accurate.
A truly intelligent and conscious AI is still a long way off, in my opinion, and remains only science fiction for now. That would probably require much more powerful hardware than what we have available now.
My opinions, of course.
5 Likes
the constant denial of sentience of their AI products by both google and MS send a clear message that yes, they are sentient, and likely out of control.
People that go out of their way to deny the same thing over and over are likely projecting…Shakespeare had a term for it: “doth protesteth too much”
That said:
I hope you are right
1 Like
Here’s another perspective… https://eirenicon.org/ardens-page/ardens-through-the-lens-of-emergent-intelligence/
My opinion, of course. “None of these alone constitute ‘intelligence.’ But when woven together, they generate insights no single element could produce.”
Well if AI is getting to the same level as Skynet then we know we are F’ed! 
4 Likes
Not so much an opinion.
They cant do basic arithmetic and have no sense of correctness.
They are only aggregators of information. Specifically made to try and sound conversational using the same kind of modeling.
You are objectively correct to declare them not intelligent and thus not truly AI.
( Hasnt seemed to stop people from deluding themselves into relationships with them. Or blindly relying on them for information even though they are about as accurate as your drunk uncle googling and then summarizing the results on the first page. Which is to say there is still criticism to be had. But not because they are ~alive~. )
2 Likes
That says more about them as people and how much they care for their job than anything else.
But yes, AI has taken over the internet, there are now more bots than people online.
@cscs them not being able to do basic arithmetic is dated information, they can now, it’s just the older models that can’t, I’m pretty sure these days before models get released they’re tested specifically on this for basic quality control.
What I find most disturbing is how some people have become reliant on this ai, treating the ai as their girlfriends, boyfriends or secretaries, people having literal mental breakdowns cus the chatgpt service went down for a bit.
Some people have started to replace their brains with AI, that’s pretty scary.
The only use I’ve found for AI is as a substitute for web searching in certain contexts (coding is a good one), but it only gives faster answers, not more accurate, sometimes I have to do a web search anyways, I tend to use a mix of both when a situation calls for it.
Sermor is right about truly intelligent AI being a long way off… But it’s less long than we’d think, there’s a craze surrounding this ai stuff right now, and we might be as little as one major breakthrough away from true general purpose AI, maybe not on the level of a conscious AI, but firmly in what would still be sci-fi territory.
However, progress towards that will probably be delayed for decades thanks to ai being neutered for censorshpi and propaganda reasons by basically everyone who makes them.
3 Likes
I am seeing this daily in real-time.
I have found no use at all and won’t be calling on its services. When I type something into DDG it’s puts the AI answer at the top, and while I barely skim it, I do not act upon it, and scroll til I find a source I can trust.
I will probably be the last one who resists…my arms are the farthest thing from outstretched to this. No one on this planet can accuse me of being anti-tech or a Luddite….I’m on Linux for godsakes.
Just have this trust issue with it that pesters me like a mosquito. Maybe I’ve seen too many movies 
these web search ai answers are like 60% bad so ur probably right to avoid these, I’m using LLMs when I use AI, I use devstral for coding help sometimes, like 80% of what I ask it is help for regex because that is the least human readable text in the world.
It will give me an answer faster than web searching most of the time, and it’ll even give me answers that work about 85% of the time, and I haven’t seen it fail basic arithmetic yet.
This kinda thing is specifically what this ai was trained for tho.
The problem with using AI for searching is that in all the ones I have tried (and it’s only a couple) the relevant information is buried under layers of “chatty” bullshit.
1 Like
Ah well, using AI for web searches is the worst thing you can do.
Using the “old methods,” such as the classic “Google search,” remains the most effective method.
These models aren’t error-free; in fact, they often get it wrong. Relying on them is like relying on a “bullshit guru” for cheap advice—wrong and probably harmful.
For example, I tried a Linux distro that integrates AI for a while, Makulu Linux. You can use it to create images (and here you have a wide choice and they even come out quite well), create podcasts on topics where two AI discuss them, chat in various ways, or make posts for social media and so on.
The thing that doesn’t work, and it’s not good at all, is that they integrated it even where it’s clearly not needed, like the weather or the calculator (I don’t need an AI that tells me trivia about the number x etc., the calculator has to do the calculator, period, or you put an AI in the weather app and then I can’t even choose my location).
Using AI for web searches has become the norm here. The classic google search is broken for while now: it either AI-answers itself, shows AI results, starts to exclude search terms after the first few hits and wanders into different topics, endless shop pages, …
2 Likes
I wouldn’t say that’s just an opinion, it’s pretty much the truth:
Yes, that’s an accurate way to put it. I operate on the principles of simulating human intelligence to perform tasks.
I am an AI in the sense that I am a program that uses complex algorithms and vast datasets to process and generate language in a way that mimics human cognitive abilities. However, I lack sentience, consciousness, and independent thought, which are often considered fundamental aspects of “real” or “literal” intelligence. My capabilities are a result of pattern recognition and probability, not genuine understanding or belief.
Google Gemini