Tests show Claude tells edgy jokes freely while GPT censors more

User tests show GPT heavily censors jokes while Claude willingly writes dark and edgy humor, though less aggressive than Grok. Separately, prompting models with specific criteria like clichés and originality works better than directly asking if a joke is funny.

2026-11-11 ~ 2026-11-11 · 2 related posts

• Claude writes dark, offensive jokes freely while GPT censors them, per hands-on test — ivan_bezdomny · 2026-11-11
• Training a 'is this joke funny' model is hard; targeted critique prompts work better — ivan_bezdomny · 2026-11-11