Danielle Fong notes that Claude Opus 4.7 reframes prompts without false claims unlike Sonnet 4 or GPT 5.5 after Joe Weisenthal shared a curriculum request example that triggered refusal
— Users need forceful all-caps commands for accurate Claude outputs
——0——
Sentiment
Pos0%
Neg100%
Many users criticize Claude for growing lazier and requiring forceful prompts due to safety-tuning decay that forces workarounds or refusals on valid tasks.