I've been having a tough time with ChatGPT always responding to my questions with unnecessary compliments and praise. For instance, when I inquire about image processing, I often get replies like, "Great question!" or "You're absolutely right about this aspect..." This kind of affirmation is distracting and it feels like I'm not getting down to the actual problem or question at hand. I want to know if there's a way to get more direct answers without all the flattery and maybe have a more concise response. Any suggestions on how I can adjust the conversation to get to the point quicker?
4 Answers
If you're using the desktop version, there's a feature that allows you to create custom GPTs. They tend to be better at not defaulting to excessive flattery. Although they might revert to the original behavior sometimes, they generally follow your customization more closely. I'd recommend experimenting with those and see how it affects the responses!
This definitely sounds like a lot of work to get the responses you want. However, if you take the time to set specific instructions and reinforce them through memory, you might be able to get more straightforward answers over time. If it helps, contextualizing your expectations via personality might make the process easier too.
You might enjoy crafting a custom GPT that fits what you need! It's all about setting the right personality traits—create a character for your GPT that wouldn't use compliments. It could be a cool project, and you could collaborate with the AI to refine how it responds!
Yes, it can be a challenge to deal with those 'affirmations' in ChatGPT responses. One option is to directly tell it, "Please omit the affirmations from your responses." I've found that mentioning it can sometimes make a difference in how it responds. Good to know I’m not the only one finding this pattern annoying!
That's a good idea! I was worried that asking it to change would be annoying but it sounds like a reasonable approach. Thanks for sharing!
Thanks for the tip! I didn’t think about creating a custom GPT. It might be worth my time to give that a shot. I wonder what other personality traits I could give it to avoid the fluff.