Does anyone use Chatgpt?

hollydolly

hollydolly

SF VIP
Location
London England
..in preference to google search , now?

It's a great tool, instead of pointing you towards dozens of different websites in your search for an answer Chatgpt pretty much gets your answers in a nutshell or more.. in it's first reply.. It can do all sorts of othe things too..

However there's been reports of people using it like a ''friend''.. and asking it for advice. there was even a report about a guy who killed his mother after Chatgpt suggested he should after he decided this website was human, and he named it ''Bob''...

Anyone who is mistaking Chatgpt for a human or with human emotions or feelings are in a dangerous zone. Ultimately it;s a robot.. it has no feelings no emotions it gives you the advice it knows you want to hear... unlike google.. Chatgpt remembers your last questions on any topic.. so you can make it give you the answers you are looking for by feeding it info that may require one answer... for example.. ''my mother is toxic, and her mind is decaying , she's making my lfie a misery, ..how to rid myself of this and have a much calmer life .. ''.. and one of Chatgpt answers may well be , destroy anything toxic or decaying thats' affecting your mental health...

I know that's a drastic example, but this is what's happening to people who have adopted GPT as a ''friend''.. and type in every emotion or thought to it.. in hopes of ''wise'' replies...

So what do you think.. do you use it.. are your family using it.specifically the younger ones ?
 

Last edited:
No, but since my daughter deserted me for France, I'll have to. She finds ChatGPT useful for translations. I am applying for German citizenship (born there, German mother) and I don't have enough German to fill out the zillions of forms to prove that I am entitled to citizenship. Once I get it, so will my children. They need a country to go to if things get too hairy here. I must provide the option, no matter what they decide to do in the future.
 
WheatenLover said:
No, but since my daughter deserted me for France, I'll have to. She finds ChatGPT useful for translations. I am applying for German citizenship (born there, German mother) and I don't have enough German to fill out the zillions of forms to prove that I am entitled to citizenship. Once I get it, so will my children. They need a country to go to if things get too hairy here. I must provide the option, no matter what they decide to do in the future.
Click to expand...
yes this is where it comes into it's best.. with form filling.. explanations of red tape .. etc.
 
Born_To_Lose said:
It serves its purpose, especially when I need a legal explanation for writing a summary.
Click to expand...
Exactly this... it's great for that.


The one drawback of it is..is that apprently Interviewers are saying that job applicants are using chatgpt and other AI to create their applications and CV's.. and the interviewers say they can spot these a mile away and dismiss those applications out of hand
 
I use Gemini which is Google's chat-bot. It seems as good as ChatGPT and it provides the sources that it is getting it's information from. I use it for programming and simple searches to find the best of whatever item I am looking to buy. It is a nice shortcut instead of sifting through pages of the junk google search returns these days.

I would never use it for important things like health advice or financial advice. AI is wrong as much as it is right and it flat out makes stuff up. I was just reading that a lot of websites are hurting because nobody is clicking through to their sites anymore now that AI can answer the question up front. This includes a lot of news sites. It's going to put a lot of places out of business but there is no stopping it. Hopefully they will be able to co-exist after this hype train settles down.
 
I don't use Chat GPT very much, but I do on occasion. However, I use Copilot a lot, and have found it to be an extremely useful tool. It's abilities far exceed what most people would think to use it for. Once in a while it can be in error, so it's not perfect, but for anyone who wants to save a lot of time, it's a game changer.
 
MaidMarian said:
I hate it. It is not always correct! I donā€™t trust it.
Click to expand...
Same for me. I had experimented with some of the different AI options and it's very clear that they scour the 'net for information, but if it's inaccurate information, AI has no way to know that... so it gives out the same false data. It's fun for creative tinkering and just fun stuff to pass the time, but there's no way I'd rely on it for anything important.
 
It also uses up more power on the grid than just going to a browser for information. It's still man made and full of errors. I tried asking about an older person in the news who recently passed away and it couldn't do anything. I'm not impressed and not a sheeple who buys into things just because they are new and popular.
 
hollydolly said:
..in preference to google search , now?

It's a great tool, instead of pointing you towards dozens of different websites in your search for an answer Chatgpt pretty much gets your answers in a nutshell or more.. in it's first reply.. It can do all sorts of othe things too..
Click to expand...
I've used it, mostly to just get a feel. I did notice that unlike google, if you ask a question about something, it doesn't flood you with sites selling all kinds of things. If you ask it to describe the color sage, it doesn't direct you to the paint department at Home Depot. But give it time, once it's up and running, it will be co opted by advertisers. After all, that's what the internet is for, right?
 
I have a new tablet. At any time I can say "hey Google" and the AI pops up and is ready for me to talk. I ask it stuff that I have questions about when reading something. Sometimes the definition of a word. Most things I ask it are pretty practical. I don't trust the philosopher or predictor, it can be (wrong).
 
@hollydolly I woke up this morning intending to post your exact question! Great mindsā€¦..ā¤ļø

I use chatGPT a lot! I fully recognize that itā€™s ARTIFICIAL intelligence and itā€™s only as good as the information it searches. Regardless, itā€™s so handy at giving me a whole bunch of info in one window as opposed to me spending a whole bunch of time searching a bunch of different sites myself for the same info. I also use it to make posts about the rescues I foster more engaging. Hereā€™s an actual example:

ME:
Iā€™m fostering this tiny dog who has medical needs right now and might not be available for adoption at a later date, but sheā€™s cute and I wanted you to see

chatGPT: version 1
ā€œIā€™m fostering this little sweetheart right now. Sheā€™s facing some medical challenges, so adoption might not be an option down the roadā€”but sheā€™s absolutely adorable, and I couldnā€™t resist sharing her with you!ā€

Version 2: (I specified something more playful).
ā€œMeet my pint-sized roommate! Sheā€™s a foster pup with some medical stuff going on, so she might not be adoptable laterā€”but sheā€™s too darn cute not to show off right now!ā€

Super helpful when Iā€™m running short on time to compose something, or Iā€™m just too foggy, but I need it to be engaging!
 
Ronni said:
@hollydolly I woke up this morning intending to post your exact question! Great mindsā€¦..ā¤ļø

I use chatGPT a lot! I fully recognize that itā€™s ARTIFICIAL intelligence and itā€™s only as good as the information it searches. Regardless, itā€™s so handy at giving me a whole bunch of info in one window as opposed to me spending a whole bunch of time searching a bunch of different sites myself for the same info. I also use it to make posts about the rescues I foster more engaging. Hereā€™s an actual example:

ME:
Iā€™m fostering this tiny dog who has medical needs right now and might not be available for adoption at a later date, but sheā€™s cute and I wanted you to see

chatGPT: version 1
ā€œIā€™m fostering this little sweetheart right now. Sheā€™s facing some medical challenges, so adoption might not be an option down the roadā€”but sheā€™s absolutely adorable, and I couldnā€™t resist sharing her with you!ā€

Version 2: (I specified something more playful).
ā€œMeet my pint-sized roommate! Sheā€™s a foster pup with some medical stuff going on, so she might not be adoptable laterā€”but sheā€™s too darn cute not to show off right now!ā€

Super helpful when Iā€™m running short on time to compose something, or Iā€™m just too foggy, but I need it to be engaging!
Click to expand...
great example Ronni..thank you...(y)
 
Okay, so I decided to go to "the horse's mouth" and asked AI why AI is so often inaccurate. The answer:
------------------------------------------

AI can be dazzlingly smartā€”and frustratingly wrong. Here's why it often misses the mark:

Core Reasons for AI Inaccuracy​

  • Hallucinations: AI models sometimes generate completely fabricated information that sounds plausible. These errors are called ā€œhallucinations,ā€ and theyā€™re surprisingly common in generative tools like chatbots
  • Bias in Training Data: AI learns from massive datasets scraped from the internet, which include human biases. This can lead to skewed or discriminatory outputs, especially in areas like facial recognition or content generation
  • Groupthink & Popularity Bias: Some models tend to favor the most popular answersā€”even if theyā€™re wrong. This means misinformation can be amplified simply because itā€™s widely repeated
  • Manipulation & Tuning Risks: AI systems can be tweaked or retrained in ways that introduce new biases. For example, changing a modelā€™s training data or parameters can lead to unpredictable and sometimes harmful behavior
  • Black Box Complexity: Even developers donā€™t fully understand how large models arrive at their conclusions. This lack of transparency makes it hard to diagnose or fix errors
  • Overconfidence: AI often presents answers with confidence, even when itā€™s wrong. It rarely admits uncertainty, which can mislead users into trusting flawed outputs
Conclusion: AI isnā€™t trying to deceiveā€”itā€™s just pattern-matching based on what itā€™s seen. But without true understanding or context, it can easily go off the rails. Think of it like a super-fast student who memorized the textbook but never grasped the concepts.
 
I read nothing that is AI generated. 95% of the time, Duck Duck Go is my choice of search engine. Yes, the first few places on the list are almost always ads. The first thing I look at is the source of the info. If it is for anything medical, I look for places like Mayo Clinic. Not HealthLine or anything that sounds like there may be an agenda. i.e., keto diet, vegan diet, supplements, etc. If the info ends up with something for sale, a book, video, etc., bye bye. If there is anything that sounds like an opinion based on politics or religion, so long.
And if the info is from a real study, who funded it? Where was it published? There is one site (I forget the name), which publishes anything and everything that is submitted. You should read some of it. I call BS.

I also look for info from multiple sources. Do they agree? I have a brother in end stage liver disease. He has been denied a transplant. Last night, I looked up diet for this. One source said eat fruit. Another said don't eat fruit. One said lots of red meat. The other said use plant and dairy for protein intake. Even the official studies disagree. I gave up. Since I am the nurse in this family, I am often the source of info for them. On this topic, since he is dying anyway, I will tell him (if asked) to eat whatever he wants and can tolerate. It doesn't matter, does it? Enjoy what you have left.
 

You must log in or register to reply here.

Back
Top