3 ms·

I agree with this, but also get so irritated with the endless praise. Like, tell me I'm wrong and tell me why. If I present a bad idea, tell me that. Or even

I agree with this, but also get so irritated with the endless praise. Like, tell me I'm wrong and tell me why. If I present a bad idea, tell me that. Or even like "huh? explain" Looking forward to ramping up the honesty parameter. Hoping OpenAI's new voice model trivializes this so I don't need to prompt engineer.

fassssst 2y ago
Curious what happens if you set custom instructions to correct that. Does it work?
pr337h4m 2y ago
A simple "Be brutally honest" at the end works surprisingly well.
vundercind 2y ago
Me: “I’d like to start a business harvesting cat shit to fight global warming. People would pay me to take their cat shit and bury it because maybe it has carbon in it or something IDK I’m not a scientist. I would give them carbon credits in the form of a cat-shit-sequestration crypto token my business created. I’d pick up the cat shit by bicycle to save carbon emissions. Please help me develop a business plan.” Chargpt: “that’s a great idea for a business! Here are some suggestions…” Not a real exchange, but god, it does feel that way sometimes. One reason you can’t trust the damn thing is it’s too positive.
- satvikpendem 2y ago
  I just used your prompt on GPT 4o appended with "Be brutally honest, if the idea is bad, feel free to let me know without any sugarcoating" as sibling comments have suggested, and it works pretty well and doesn't give false platitudes.
  - vundercind 2y ago
    Nice, will start using that. … it is still a little weird that the default is, like, psychotic levels of positivity, though.
    - satvikpendem 2y ago
      The creators of these LLMs train it not to offend, to an extreme degree, so much so that it sounds unnatural to humans.
- kedean 2y ago
  Tried putting that verbatim into Phind, got back an essay on how to get funding to start a business. In the conclusion was this gem: "While the idea of harvesting cat feces for carbon sequestration is innovative, success will depend on thorough research, strategic planning, and effective execution."
- richardw 2y ago
  I imagine the first models that trained on HN comments resulted in staff members resigning and becoming hermit alcoholics after some brainstorming and "Show HN" test runs. To make something that didn't result in instant and persistent despair likely meant a lot of sentiment analysis on input data, and the result is a system that is extremely pleased every time we show up.
hobobaggins 2y ago
That's a truly innovative approach! This will create a robust solution to that problem.

fassssst 2y ago

Curious what happens if you set custom instructions to correct that. Does it work?

pr337h4m 2y ago

A simple "Be brutally honest" at the end works surprisingly well.

vundercind 2y ago

Me: “I’d like to start a business harvesting cat shit to fight global warming. People would pay me to take their cat shit and bury it because maybe it has carbon in it or something IDK I’m not a scientist. I would give them carbon credits in the form of a cat-shit-sequestration crypto token my business created. I’d pick up the cat shit by bicycle to save carbon emissions. Please help me develop a business plan.” Chargpt: “that’s a great idea for a business! Here are some suggestions…” Not a real exchange, but god, it does feel that way sometimes. One reason you can’t trust the damn thing is it’s too positive.

satvikpendem 2y ago
I just used your prompt on GPT 4o appended with "Be brutally honest, if the idea is bad, feel free to let me know without any sugarcoating" as sibling comments have suggested, and it works pretty well and doesn't give false platitudes.
- vundercind 2y ago
  Nice, will start using that. … it is still a little weird that the default is, like, psychotic levels of positivity, though.
  - satvikpendem 2y ago
    The creators of these LLMs train it not to offend, to an extreme degree, so much so that it sounds unnatural to humans.
kedean 2y ago
Tried putting that verbatim into Phind, got back an essay on how to get funding to start a business. In the conclusion was this gem: "While the idea of harvesting cat feces for carbon sequestration is innovative, success will depend on thorough research, strategic planning, and effective execution."
richardw 2y ago
I imagine the first models that trained on HN comments resulted in staff members resigning and becoming hermit alcoholics after some brainstorming and "Show HN" test runs. To make something that didn't result in instant and persistent despair likely meant a lot of sentiment analysis on input data, and the result is a system that is extremely pleased every time we show up.

hobobaggins 2y ago

That's a truly innovative approach! This will create a robust solution to that problem.