How many hours per day do you use chatgpt, gemini, etc?
Is anyone addicted to AI?
Frankensnore on 03/03/2026 - 21:39
Comments
- 1
- 2
I used to think the same about things like Instagram, Twitter, and TikTok yet here we are today, "people" have proven me wrong lol.
Side rant is I wonder if they would still be just as addicted if they realised the cost impact on stuff they want to buy like computing and gaming devices (if not all the other stuff like energy, water, etc.).
I don’t understand how people can use social media for hours straight. Shit is so boring after a few scrolls. It’s so repetitive
Call me old fashioned but AI will never be able to replicate authentically the level of organic mspaint workmanship as seen here
Challenge accepted
Gday lumo !
Pretty cheesey
… idk what I’m looking at but I will never look at mac n cheese the same
🤮
Because naked people wrestling in red jello is so last season.
Not gonna lie… the kidpix ones were always my faves. :D
AI means I can generate these masterpieces of jv in seconds
You and me both. Once you learn how, it's like riding a bike
Quite the opposite.
Per day? I would say zero… per month, maybe 2 to 5 mins while it makes a stupid picture for me to laugh at…
it makes a stupid picture for me to laugh at
what prompt did you use for that? its never made anything funny for me
Make a video of uncle Ian, dressed in his Sunday best overalls, working on a Chrysler Valiant in his rundown workshop while looking confused about all the enquires he gets from OzBargain and keeps asking “who are all these cunce?” as he puffs from a cigarette that seems to be permanently attached to his bottom lip”
@Wiadro: Ohhhh, I see that Uncle Ian's much hotter business partner or neighbour is on the tools today. Im Not climbing that until he has a bath with soap and water though. That's just nasty.
I take it that Rolts has persuaded Uncle Ian to be the note speaker at this year's Woke ShiT Box Convention? I think the topic is about how to manage inflation when putting air into tyres of woke shit boxes
Cunce… Nice
If you're asking the question then you're probably using it too much.
https://www.media.mit.edu/publications/your-brain-on-chatgpt…
https://theconversation.com/mit-researchers-say-using-chatgp…
AI induced pychosis on people who had NO history. YIKE https://www.youtube.com/watch?v=MW6FMgOzklw&t=352s
Honestly barely at all. I don’t want to contribute to its pollution/water usage/world domination.
But really I also want to continue to use my own brain and unless my own ability or googling has really got me stumped that’s really the only time I might go to AI.
It's being heavily pushed by my work (large global IT Company), and I use it sparingly (copilot sigh) for some troubleshooting. I'm not a coder though; more ops/infrastructure etc.
I dislike that everyone is using it to write emails/summaries for them, zero effort can be seen often and summaries often miss context. It's very obvious when middle/upper management are using AI and when I'm being asked to review/proofread someones AI email (from someone 2 ranks above me) it's pretty bad..
I don't mind AI as a TOOL; but I don't like that people are using it to replace common sense, their own critical thinking, and their ability to function in the workplace. It should just be another tool/extension of yourself to make your life easier/maybe something more efficient about something you already know. It becomes very obvious when those people using AI to generate their docs/emails then have to speak to it in a meeting/get asked questions about stuff and they fumble.
My workplace is pushing this heavily too. My first thought is that AI can replace most of middle management quite easily.
Haha 100%! They're the most replaceable for sure but it never happens that way.. low level workers always first sigh
You can't really use AI for a lot of work comms because the material it produces is intrinsically slop and audiences see straight through it.
As an example I was asked to produce a super generic piece in super generic terms on a particular topic to advise an audience. Perfect for AI, right? I asked AI, ended up having to spend another hour basically replacing all the content to make it concise and meaningful. The outcome? I was asked not even to present it.
Next meeting with same audience, I spent 15 minutes only writing it up myself with no AI, my content was concise, relevant advice and I got very positive feedback.
Same with job applications - I've never felt any success using it for that either.
At the end of the day, if people can't extract a meaningful message from your communication, it goes in the rubbish bin .
It is getting a bit ridiculous where I work:
We are told we have to use AI to write our own performance reviews. Managers are told they have to use AI to summarise said reviews.LOL WHAT?!
Nah get me outta here. I'll be in the countryside tending to chickens soon enough ffs.
It's quite useful as a tool for work, especially when you setup custom agents for specific subject matter areas. Don't use it in personal life though. Haven't found a need to beyond the summarised searches.
The trouble is that AI sounds authoritative, but it makes stuff up.
Yes, and you notice how bad it is when it's a subject you already know a lot about.
The trouble is that AI sounds authoritative, but it makes stuff up.
Yeah AI will never say 'I don't know' because it literally does not know anything, it just generates a response to any prompt - so even when it's correct it didn't actually 'know' the answer lol
If you tell it "do not give me any response without looking up the correct answer first" then it does a lot better - because it goes off and googles the answer then relays it to you.
@trapper: I go one step further and say "could you please provide me with a relevant link to support your information?"
I go one step further and say "could you please provide me with a relevant link to support your information?"
ChatGPT: here's a link I hallucinated it will give you a 404 error or simply not contain the information but it supports the information I gave you :)
@Muppet Detector: And provide confidence level for each response.
AGain, AI is a tool, proper prompt engineering is needed to get the best - using it like a search engine is not using AI models right.
@trapper: @trapper That's not even accurate because the 'source' could be any random reddit post, sometimes even from someone making a joke or saying something sarcastically.
@Vool: It will cross reference multiple results and look for a primary source.
It's definitely more accurate that generating 'likely' next words for a topic in was never trained on.
@trapper: It doesn't though. In fact yesterday I had a co-worker come to me in a panic saying there's a new Mac OS update that's killing Macbooks, Had a look around and couldn't see anything so asked where he saw it and he said a Google AI result. I told him to check what it was sourcing and it's source for this 'new update' was a Reddit post from 18 months ago where a single person had a Macbook die after installing an update. Not only was it's source not relevant, it was way out of date.
Sounds like politicians?
Ask it to not sound authoritative when framing answers in the settings then.
@StiffHindQuarters: My self-driving car keeps ploughing through stop signs!
Then just disable the "ignore road rules" setting, what's the big deal
The trouble is that AI sounds authoritative, but it makes stuff up.
You sound authoritative.
@Sal in SA: I have had AI link its proof to me and somehow it misinterpreted the link and was just plain wrong.
@Sal in SA: Chatgpt 4. I was also getting it to do currency conversion calcs that required to go from USD pounds weight, to AUD KG weight at different currency conversion rates. It was doing mostly fine until it just got to a certain currency level then it decided to change how it calculated things. It didn't know it was doing it wrong even after I explained how it might be calculating things wrong. It didn't fix it, it just excluded it from its results. After that, I couldn't really trust it to do even more detailed calculations that require more steps as it might get to a point where it is slightly off but without calculating everything to check it, I might not know
@serpserpserp: a lot has changed in that time - I mean a lot - try ging claude.ai a go for comparison now - it will be a very very different experience.
@Sal in SA: Can only use enterprise copilot unfortunately. Not sure what LLMs it has access to but it is limited
@serpserpserp: If you understand its limitations, frame your questions on where you want the search to go, it's good all the time. It's a language model, most people should be smarter than it, it just fills in gaps.
I changed my search URL to remove the AI nonsense that appears before the actual results, though granted google search results are also far from accurate.
Googler beware.With AI though the number of ways you can shape a search is limitless
AI != google
false equivalency. You're thinking driving a car is the same as riding a horse, and it's not.
i hope you're fact checking everything it tells you ,but i'd bet money you aren't.
I'm against AI for all the normal reasons so I don't use it for anything. I really hate that it's made all of the search engines almost useless. Not only is finding things extremely more difficult but every site (in the last 3 or 4 years) is just the same recycled trash. I'm just hanging out for when all the free trial periods end and people will go back to writing things on their own.
According to that diary of a CEO guy 99% of us will be unemployed in a few years due to AI. I'm not sure who is then buying the products of those who've sacked everyone tho.
Surely there's more to him than just another person just monetising fear using social media?
Psssst
The free use will never stop. It will be monetised just like Google search. Sadly, Google searching effectively peaked like ten years ago. Free AI searching is peaking right now, monetisation and bias won't be far off.
I'm just hanging out for when all the free trial periods end and people will go back to writing things on their own.
Unlikely.
~$35/month for virtually unlimited AI writing isn't much when you're making money from it.
An internet used to exist where people would make things for fun and not only to make money. This mindset is a big reason why everything stinks like shit now.
People got to pay the bills somehow. Housing is twice the price it used to be. And unemployment will be increasing soon due to AI taking people's jobs.
Aust governments (not just one) have set us up for a stinker over the next few years.
Allowed their lax housing policies to push housing prices through the roof right before AI wipes out jobs. So forward thinking.
None
maybe once or twice a week for personal use
maybe 100x a day for work purposesSame
If you're addicted to Large Language Model (LLM) AIs, you're simply addicted to hearing more of what you want to hear. They are descended from software like the predictive text on your phone. To see it in action on a small scale, bring up your phone keyboard, open a new email or SMS, and press the first recommended word over and over and over. Drawing from your past usage, it will produce a combination of content you are most likely to use.
Now imagine this applied on a vast scale with all the world's digital content as its database. You enter a query and at lightspeed it generates what you are most likely to want from that database, one word after another. If you ask it, it will go deeper and deeper and deeper, a layer at a time, incorporating the information you have given it. It's not thinking, it's not intelligent, it's just using probability.
Therefore if you have an ongoing conversation with it, you are merely producing predictive text in a direction you desire. That's why (if there are no guardrails or they're bypassed) people wanting to commit suicide who ask the AI whether they should do it are ultimately told how to kill themselves. That's why if you simulate war between two AIs to see what will eventually happen, they will reach nuclear annihilation. And that's why, spending time with these LLM AIs, you're getting more and more of what you want. It's just predictive text.
AI generally only acts as an echo chamber when it comes to things that fall within the realm of human opinion—and even then only to a limited extent. It typically won’t agree with you if the opinion is clearly unreasonable.
As an experiment, I’ve tried pushing an AI system to agree with me and refute robust, well-established scientific evidence, and it refuses to do so.
I expect most AI systems behave similarly (unless they are malfunctioning or hallucinating).
And I expect this tendency will become more reliable over time.
None. AI is such a useless waste of resources. Making people stupid, making electronic components more expensive due to AI farms sucking up all the ram and storage on the market. People are being forced to use it in work environments just so employers and management can justify their investment. Can only hope the bubble burst sooner rather than later. Sure it has some place in the world like things that are important like trying to find cures for diseases and answers to things that are actually important, but the amount or AI generated crap on social media is mind boggling and for what? So someone can earn a few bucks from views. Money well spent…lol
Making people stupid
This is the whole point. That, and making people depend on it (by shoving it down our throats for free/cheap currently). Once it is supposedly indispensable, they will start trying to charge for it aggressively.
At some point something has to give and it will probably topple the economy in doing so. Oops.
The collective loss of critical thinking is seriously concerning.
No
Too much enshitifcation
The dumbest person I know habitually pulls out his phone to ask ChatGPT the most basic of questions. He's offloaded his thought processes to it entirely.
I'm talking questions like "should I buy bread before the weekend in case I need some".He's not intellectually disabled either - just very mentally lazy - and when ChatGPT came along he was a moth to a flame.
There's a worrying hivemind-esque future in store for us I think, where a not insignificant portion of society will ask a chatbot to guide their worldview - politically, ideologically etc. Especially as mass media loses its efficacy.
I remember there was a study published I think a little before the 2016 US Presidential Election which demonstrated that search engines could nudge and change votes / opinions just by presenting results, e.g. politically themed news articles, in a particular biased order. I expect special interest groups can do much much more targeted and insidious manipulation of thoughts and opinions with "AI".
100% - these algorithms have been tuned for years. And with AI being a "black box" with zero transparency on its thought processes, there's no way to prove it
Negged by your friend, or their ilk…
haha, probably.
Don't get me wrong - there's a usefulness to these AIs - but they have significant dangers to the most vulnerable of society
He's offloaded his thought processes to it entirely.
Hey ChatGPT, should I go to the toilet now, or should I wait another 5 minutes?
ChatGPT didn't reply… is it too late? Maybe ask it if you should clean yourself up.
No, I'm not addicted to AI..
I use it as a tool at work though (paid CoPilot and GitHub CoPilot licenses) - it has been very useful for a non-developer IT professional to 'vibe code' complex PowerShell scripts.
ChatGPT Plus used personally, going to try going back to the free plan though.
AI is great when it works, incredibly stupid when it doesn't (so confidently wrong). Some level of critical thinking is required to properly use AI - it does you no good to believe/trust everything it says and does. It's not replacing humans in a hurry and a lot of businesses that saw it as an opportunity for mass layoffs are now facing the consequences..
and a lot of businesses that saw it as an opportunity for mass layoffs are now facing the consequences..
Hopefuly one of those consequences is humans not wanting to work for them and humans not wanting to do business with them.
AI is great when it works, incredibly stupid when it doesn't (so confidently wrong). Some level of critical thinking is required to properly use AI - it does you no good to believe/trust everything it says and does.
This is so true and very good advice.
whats AI?
Artificial Insemination. My neighbour does it with his Angus cows.
That's bull
R v Houghton (1960) tried it with dairy cows…
My neighbour does it with his Angus cows
Report him to PETA or RSPCA.
Report him to PETA or RSPCA.
He reported his neighbour for makin' bacon, but the accusations against Bradley O'Reilly were dropped and the incriminating videos have been remoooved from the Internet.
- 1
- 2
Hours?