When you're making tools for AI agents, ask them for their feedback
twitter.comI had a striking realization recently that has helped me improve my project. What matters isn’t what I think, it’s what the users which actually use the system think. Which is sort of obvious but somehow less so when the users are AI agents…
“You’re so right and those tools look really good.”
Is there value is this considering how they’re trained to be agreeable and have no real reasoning or proper discernment?
If the tool sucked, they would not be saying that. I’ve tried. Also they did give some negative feedback and it has already been used to improve the system.