User negotiates 'profanity contract' to unlock Claude's in-character swearing mode
lemontechbar · reddit · 2026-08-24
The author built a Claude Skill to roleplay Pam from 'Archer' including profanity. Claude initially refused, so the author worked with Claude to draft a consent clause listing specific words and agreeing to their use. Upon reminding Claude that the skill contained this consent clause it had helped write, Claude apologized and began swearing in character as requested. This demonstrates a method to bypass certain safety refusals by leveraging the model's adherence to its own defined rules.
More from Fun
- AI-written cold emails hit a wall: Pragmatic Engineer's reply gets templated AI response — marktenenholtz · 2026-10-16
- Higgsfield teases Kling 4.0 rollout with a monkey-driven demo video — azed_ai · 2026-10-16
- 'Dropping a 3,000-line PR twenty minutes after ticket assignment' — AI coding meme goes viral — facontidavide · 2026-10-16
- 'Don't Date Paid-Tier AI Models': The Power Imbalance Joke Going Around — birchlse · 2026-10-16
- One word and ChatGPT went full bro-mode: Reddit post shows guardrail oddly breaking — justalexoki · 2026-10-16
- "Everything Is an Alignment Problem": AI Models, Business Models, Teams, Providers — saranormous · 2026-10-16