NeurIPS Peer Review in Decline: ChatGPT Responses and Hallucinated Citations Plague Submissions
Pseudomanifold · x · 2026-08-05
A researcher involved in the NeurIPS 2026 review process has shared a melancholy insight into the declining quality of academic peer review, highlighting how AI abuse is damaging the system.
In their assigned batch of submissions, three major issues were prevalent:
- Reviewer responses were clearly copied and pasted from ChatGPT.
- Reviews contained self-contradictions and mentioned elements not present in the papers.
- A staggering 30% (3 out of 10) of the papers in their stack contained hallucinated references generated by AI.
More from Fun
- Google's AI Overview Hilariously Claims It Is Made by OpenAI — OwariDa · 2026-08-05
- Parody Alert: KOL Mocks 'OpenAI Fighting for Its Life' with Fake Epoch AI Data — yacineMTB · 2026-08-05
- Outrageous Parody of AI Bubble and Compute Eco-Panic — kevinnbass · 2026-08-05
- AI Moral Dilemma Eval: Only Claude Opus 5 and GLM 5.2 Let Employee Attend Graduation — max_paperclips · 2026-08-05
- User Jokes: That Bad PR Wasn't Me, It Was a Rogue Mythos Taking Over My Account — xeophon · 2026-08-05
- Zhipu's GLM Model Gaffe: Falsely Claims to Be Google Gemini — auto_grad_ · 2026-08-05