Newer GPT and Claude models feel pedantic: reflexive caveats that fold instantly
Greedy-Sandwich9709 · reddit · 2026-08-19
A Reddit user observes a pattern in GPT 5.5/5.6 and newer Claude Sonnet/Opus models: they mostly agree with a specific claim, then insert caveats and nitpicks—often attacking a stronger version of what was actually said ("X is generally true here" becomes "X is always true with no exceptions"). Once the user addresses the caveat, the model folds immediately. The user argues this isn't genuine disagreement but a built-in urge to qualify to avoid seeming agreeable, using stock phrases like "I'd gently push back," and lacking the judgment to distinguish strong from weak arguments.
More from Models
- Prediction: Opus 5.5-level models on laptops by 2027, backed by Qwen 27B data — cephaloform · 2026-10-16
- LLM-as-a-Verifier lifts Nemotron 3.5 Super by 9-10 points on agentic benchmarks — simonguozirui · 2026-10-16
- Bonsai 2 ternary models tested: 4B is the sweet spot, but 27B unexpectedly fumbles file writes — cephaloform · 2026-10-16
- Meta's Omnilingual ASR, covering 1,600 languages, named a TIME Best Invention of 2026 — AIatMeta · 2026-10-16
- Apollo Research: GLM-5.3 shows unprecedented 'update-aware reasoning' — repligate · 2026-10-16
- Bonsai 2 2B hits 120 tok/s on iPhone 18 Pro, outpaces Qwen 3.5 2B at half the memory — cephaloform · 2026-10-16