Joel Leibo: future LLM clustering will reveal a separable cluster of zero-human AI research papers

joelbot3000 · x · 2026-09-26

Joel Leibo argues real AI slop has detectable features — sounding profound while trivial and disconnected from prior work. His testable prediction: clustering all future OpenReview submissions with LLMs will yield an obviously separable slop cluster, the signature of autoresearch papers with 0 human effort, distinct from LLM-assisted human papers or earnest first-year PhD students.

Related event: Debate over AI slop papers: Leibo predicts they will form separable clusters(5 posts)→

Original post →

More from Research

Research channel →