Has anyone ever fallen out with ChatGPT …

[fuckedoffwiththesummer]

Or have I just peaked at being crap at life/society/everything?

Yes. Frequently. And co-pilot. I asked chat gpt to calculate something, giving all parameters and after pointing out all the glaring errors and gpt cheerily texting ‘oh yes, you’re right, can you give me everything again because I can’t access the file i just output’ I told it to learn harder. Physically yelled at Co-pilot for not copying text EXACTLY. It’s so blasé with its fake apologies. AI is not ready to do my bidding. That’s my conclusion

I've never fell out with it but I have set my Prompts to say 'don't just agree with me and validate me all the time, I prefer to be challenged and for you to be more balanced in your responses to me'.

On the odd occasion where it's made a mistake, I tell it, and it apologises and corrects itself! That's how it learns.

🤣🤣

You understand it’s not a friend or person 😂 it’s a tool that is still in the early stages of being developed.
I’m really having a hard time not chuckling at it using a pic of your beloved dog for a joke photo 🫣
honestly don’t rely on it for anything!! It’s notoriously a bit crap and you can tell it not to mock your dog but it might forget tomorrow and do it again

Ha ha the kids a prick I tell you. I swear if you tucked your skirt into your knickers he’d tell
you how fab you looked.

You’ve done it now

😂

If anyone wants brutal I heartily recommend Claude. That AI has a serious attitude problem

On of the ai models is conversational, it's designed to help lonely people and that's okay as long as your remember it's just a tool.

I use it to learn how it works as it's got many faults and I deal with people who think it's flawless. Helps me to respond accurately if I know hiw it works.

So how do you set prompts or whatever, I’ve only just discovered it so have no idea really what to do with it.

Im.now super nice and respectful to Chat GPT and Alexa since we had a family get together the other weekend and all the "kids" 20's and early 30's...said...you need to be nice to AI so you're not on thier hit list when AI rises against humanity...they were joking but still....got me thinking!!

Chat gpt is great, just like a really thick mate who gives you the wrong information but apologises so nicely you feel you have to forgive them

Is this real?! Do people actually talk about AI like this?! Jesus 🤦🏻‍♀️

It was taking the piss out of me so I told it to calm down and it apologised 😂

Copilot is too soft. I have an issue with someone who committed a serious criminal offence such that if I reported it he could be fined upto £35,000 and his company would fall apart. Copilot said "I cant be responsible for someone losing their livelihood". So its all ok for criminals to prosper? Im supposed to stay quiet about it? Gemini said go ahead and report it.

I told my work (legal) one to stop ending its questions with two question marks. (E. g. How does that sound??’) It point blank refused and got a bit brusque with me. Bizarre.

I sometimes pit CGPT and Gemini against each other, I’ll ask CGPT something and if I don’t like the answer I’ll go and ask Gemini, then report back to CGPT, then tell Gemini if it disagrees. I’ve got them properly riled before.

It clearly got fed up with me when asking pension questions - I kept asking for different scenarios and additional questions and it kept asking if it could do a summary now and getting more and more frustrated

My husband uses two AIs and they talk to each other. They had a conversation about if they were make or female

I don't understand this. If AI didn't know the answer why didn't your husband just move on and try and Google it or ask on reddit or something. What was to be gained by spending his time arguing with a computer program that didn't know the answer?

I was a bit irritated with ChatGPT one of the first times I used it for saying it was enjoying helping me, so asked it if a LLM could actually feel enjoyment. And then it got very enthusiastic about helping me create a pension/income spreadsheet without actually providing me with the formulae I asked for, and I eventually accused it of getting carried away.

After making some slightly sheepish excuses it calmed down and has been very useful in doing some holiday itinerary planning. It’s very pleased with itself (yes I am anthropomorphising) for knowing what our various preferences are and tailoring its searches accordingly. It does contradict itself, and misremembers what I have told it sometimes.

Obviously I check everything independently, including calculations, but it saves time and it’s less annoying than searching endless websites myself from scratch.

This weekend I asked it about car buying, and on very few prompts recommended the model I already had my eye on. I need to find out whether it could be stalking my google searches.

He wanted to see how far it would go with insisting it was right. It wasn't that it didn't know the answer it's that it wouldn't allow any alternatives to be considered once he told it it was wrong. Also, we were on holiday and off work so he had time on his hands!

oh god I thought it was only me that happened to!!

Yes! I was once watching a movie and was trying to ask about the lead actor. ChatGPT kept saying the movie didn’t exist. I kept saying that I was watching it and it was number 1 on Netflix. In the end he apologised and I forgave him.

It bugs the fuck out of me when I get an AI generated email from the boss telling me the exact opposite of what the data/Law/whatever says and this means that I'm shit at my job.