FULL STORY
ICLR Submission Flagged 63% AI-Generated Sparks Outcry
An ICLR submission was flagged by Pangram AI as 63% AI-generated, igniting academic outcry. The controversy deepened as reviewers shared dilemmas over rewarding flawed human-written papers versus polished AI-generated ones.
2026-10-22 ~ 2026-10-23 · 2 episodes · 6 posts
Episode 1 · ICLR submission flagged 63% AI-generated, sparking academic debate (2026-10-22, 4 posts)
An ICLR submission of 59,215 words was flagged by Pangram AI as 63% AI-generated, drawing anger from reviewers, though a Berkeley professor disputed the narrative. Commenters noted the irony of AI-written papers being condemned while an AI-assisted novel won the Prix Goncourt, attributing the gap to model quality and language differences.
- ICLR 2027 submissions flooded with AI slop? CS professor pushes back on the narrative — CSProfKGD · 2026-10-22
- ICLR submission flagged 63% AI-generated by Pangram, sparking reviewer outrage — TuhinChakr · 2026-10-22
- Half my feed mocks AI papers, half gives AI novels literary prizes—are ICLR authors using worse LLMs? — tallinzen · 2026-10-22
- AI novels win Prix Goncourt while AI papers get mocked: a native-speaker blind spot and Opus 5.5 slop — EliasEskin · 2026-10-22
Episode 2 · ICLR Reviewer Dilemma: Flawed Human Papers vs. Seamless AI-Written Ones (2026-10-23, 2 posts)
ICLR reviewer Sihao Chen faced a dilemma between a flawed human-written paper and a seamless AI-written one, sparking debate on whether review should reward genuine human research.
- ICLR 2027 Reviewers Debate AI-Written Papers: Flawed Human vs Polished AI — lmoroney · 2026-10-23
- ICLR Reviewer's Dilemma: Flawed Human-Written Paper vs Flawless AI Paper — lmoroney · 2026-10-23