Ryan Heuser @ryanheuser.com · Feb 20 AI completions of historical poems bias emotion toward positivity and away from arousal. LLMs prompted with an emotion taxonomy and a poem, for 3 taxonomies x 3K human poems [Chadwyck-Healey sampled for poet DOB 1600-2000] + 3K AI poems [9 LLMs completing first 5 lines of human poem].
20 likes 5 replies
?
Replies
I want only this in my timeline, please!
that's cool! I wonder if abstract lexis is partly responsible for the positivity bias
Surely the point is that the completion lines are basically trite and generic. The positivity comes from lack of complexity and nuance—two of the keys to good lyric poetry. AIs are shit poets.
What emotion taxonomies did you use?
Waste of water