2d ago

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
Original post

Claude recently feels even more lazy than usual. I have to force him to do his job properly and correctly. At some point, I'm just writing in all caps and yelling at Claude in my head. Dont know if it helps.

9:33 AM · May 18, 2026 View on X