Debate Erupts Over Who Really Invented RLHF

A debate on X between binarybits and andrewprock questions whether OpenAI invented RLHF, with a 1993 robot shaping paper cited as early evidence that the core idea predates the 2017 work.

2026-10-11 ~ 2026-10-11 · 2 related posts

• RLHF predates the acronym: a 1993 robot shaping paper already described human trainers — andrewprock · 2026-10-11
• Was RLHF really OpenAI's invention? A debate over the 2017 precedent — binarybits · 2026-10-11