Student finds Opus and ChatGPT trash his slides, then reverse after seeing the source material
Kazoru4 · reddit · 2026-09-16
A student describes using NotebookLM section-by-section to summarize textbook readings and build presentation slides — satisfactory but with odd mistakes. When Opus 5 and ChatGPT 5 Luna reviewed the results, they trashed everything as wrong and lacking causality; after he uploaded the source material, they immediately reversed and confirmed the slides were factual, then resorted to nitpicking trivial details to seem smarter.
He concludes that newer models "used to hallucinate by agreeing too much, now they gaslight by nitpicking," questions the "AI is too dangerous" narrative given how dumb LLM reasoning still is, and argues the development slowdown makes sense — but because saturated data is making models dumber and a different kind of progress is needed.
More from Models
- Mozilla Report: China-US AI Model Capability Gap Narrows to 4.4 Months — External_Mood4719 · 2026-09-16
- Dev recalls fighting for structured output in GPT-3 days as dynamic inputs now parse in <200ms — blixt · 2026-09-16
- llama.cpp Merges hc Ops for Qwen4exp, Time to Re-benchmark Qwen Flash — jacek2023 · 2026-09-16
- Apodex 1.1 demos: raw clinical tables to Kaplan-Meier curves, XRD indexing, protein docking — aakashgupta · 2026-09-16
- Apodex 1.1 launches: moves from deep research to real task execution — testingcatalog · 2026-09-16
- Indie dev open-sources Qwen-2.5-1B-RLCD with 5x faster on-device JSON inference — suchenzang · 2026-09-16