Training models on human attention is a huge opportunity

daniel_mac8 · x · 2026-08-18

The author suggests a big opportunity lies in training models on how human attention works, acting as a reverse harness for how models interface with us. While LLMs possess theory of mind, they still struggle to balance multiple audiences, such as separating end user and creator needs in coding contexts.

Related event: LLMs match humans on theory of mind yet stumble with multiple perspectives(4 posts)→

Original post →

More from AGI Musings

AGI Musings channel →