7 Comments
User's avatar
Suhrab Khan's avatar

Fascinating approach! Having ChatGPT argue with itself really highlights blind spots and refines ideas. Your breakdown of Optimist vs. Skeptic vs. Neutral Analyst makes the process so actionable.

I talk about latest AI trends and insights. Do check out my Substack, I am sure you’ll find it very relevant and relatable.

Expand full comment
Mia Kiraki 🎭's avatar

WOW I love this!! Of course you had to complicate things hahah. The 3-role structure is soooo much more useful than the 2-voice version.

Expand full comment
Kurt Schmitt's avatar

Yes it is! I'm already thinking about other uses for this. I may create a custom GPT for it. We shall see.

Expand full comment
Karen Spinner's avatar

Wow, this is fascinating! The benefits of adversarial prompting are pretty intuitively obvious.

But, whenever I try it, I end up overthinking and procrastinating. 🤣

Expand full comment
Kurt Schmitt's avatar

Haha! Yes, the evil twins, overthinking and procrastination. The idea is to reduce overwhelm and decision making but if it has the opposite effect, well, that's not good!

I have a love-hate with all of this stuff because part of me wants to automate all the tedious stuff away and part of me wants to control every piece of it. Note to self: just let go... 😊

Expand full comment
Karen Spinner's avatar

I can totally see that! For now, I’m sticking with what seems to work best for me…posting whatever pops into my head.😆

Expand full comment
Kurt Schmitt's avatar

That seems to be working well for you, so keep on keeping on! 😊

Expand full comment