Professor fed his 2-year AI-companion debate with a friend to an AI and published it as a dialogue

minzlicht · x · 2026-09-16

Psychology professor Michael Inzlicht argued with a friend about AI companions for two years without either budging. So he fed everything they'd said to an AI, had it rewrite the fight as a dialogue, then heavily edited it and published it on Substack as An Argument With a Friend Who Doesn't Exist.

His verdict is a wry concession: his friend is 'usually right. This time I'll only say he's not wrong' — hinting his own opposition to AI companions may be softening. It doubles as an interesting experiment in AI-assisted writing.

Original post →

More from AGI Musings

AGI Musings channel →