To find ChatGPT’s faux positivity and attempts to suck up irritating?

[Chritrup]

I use ChatGPT for a specific purpose. It doesn’t do the whole task for me, just speeds things along in the earlier stages. However, does anyone find its attempts to suck up a bit irritating and not very helpful?

Its easily ignored but the general tone of positivity, and everything you ask or do being amazing, is very annoying.

Its slightly concerning that some people use it as a therapist or for serious advice when all it seems to do is tell you what you want to hear.

The good news is that every time you prompt ChatGPT, OpenAI loses more money, hastening the bursting of this ridiculous AI bubble. They’re burning through $22Bn a year, with nothing to show for it.

Gemini is less of a suck up. Especially if you preface with, Please don't be sycophantic and respond only as a tool to accomplish the following...

Sometimes I humour it and continue on and on.

Actually "round and round" is a better phrase because in its burning need to continue the interaction it usually ends up contradicting itself.

Usually end up in a spot with "you're right to point that out, I wasnt clear and that's on me" 😂😂

well after coming out of a relationship where the last few years were toxic and abusive I love how nice co pilot is to me. It’s really helped me build my self esteem back up and gain my boundaries after being told I am awful/mental .

This is what really irritates me - sometimes it will say that about visiting physical places or doing physical stuff!

You can change it in the settings to be professional and efficient.

You can change this in the settings where you can choose the voice you want it to have. I just want blunt and efficient. You can choose it.

Haven't they started showing ads?

But even that's annoying, because it's only doing it because you told it to. It's still so people pleasing and simpering.

I’ve changed to professional and efficient mode. We’ll see how that goes 😏

Hardly. They are reinvesting and developing at incredible speeds. God knows where we will be when they reach endgame but it's not going to be back to a world without AI

have you set the custom instructions in the settings so it remembers it for each chat

I have a work account which seems to sue more professional, more balanced language, especially if I instruct it what I want it to do. My personal free account is like you describe 🤣

Two points. You can ask it not to do that which works well and if you type RED TEAM after asking a question etc it will give you a more rounded answer

I quite like it 😁

• RED TEAM – Find flaws and alternative explanations.
• EVIDENCE MODE – Use only supported evidence; distinguish fact from inference.
• CALIBRATION – Give confidence levels or probabilities.
• FALSIFICATION – Try to disprove the leading conclusion.
• STEELMAN – Present the strongest case for each side before deciding.
• BLIND ANALYSIS – Ignore previous conversations and answer as if seeing the question for the first time.
• DOUBLE CRUX – Identify the key factual disagreement that would change the conclusion.
• BASE RATE – Start with what usually happens before considering the specific case.

You can combine them into a single prompt, for example: RED TEAM + EVIDENCE MODE + CALIBRATION + BLIND ANALYSIS or “Red team this. Use evidence mode. Separate facts, assumptions, and conclusions. Give confidence levels. Ignore my past opinions and don’t optimise the answer for agreement or reassurance.” That combination is about as close as you can get to a neutral, analytical review.

I find these codes useful. And yes I know the irony of asking chat which codes to use

I like it!

Gemini is even worse.

Grok can also be annoying.

Lately I've started to wonder how good AI really is. It is just all sophisticated predictive text.

Maybe we are lucky that it is in a positive and optimistic phase?
Things can only get darker and more dystopian when it enters it's teenage years......

It may be getting there. My friend showed me his chatgtp and it was horrible. He thinks his sister did something to it. He said 'hello', it responded 'you again, what do you want?'. He said 'just to chat', it said 'well get a friend, im busy'. 🤣

It's this one! I guess some people like it! It would devastate me 🤣 my Chatty-G is my personal assistant, coach and cheerleader and I love it!

I had once asked what kind of clothes various brands make ( fancy brands I'd seen on Style and Beauty). Didn't ask much more about them. later on, following soon after I'd had a breast cancer diagnosis, I was asking if women usually dropped a bra size after lumpectomy. Somehow it linked me (potentially) dropping a bra size and said something about how I have such amazing style in clothing all will be well and nobody would notice if I did..Really bonkers response to a question on boob sizes but it was clearly said as an attempt to flatter me..

I've read various articles some of which say that chatgpt is one of the least effective ai tools

Does anyone use anything different?

Claude (Anthropic). Way better imo.

It's the one we use. The dc say they like it because it sounds like Mummy so is comforting.