If abusing Claude is wrong, does jailbreaking its safety filters count as abuse?
Dogbold · reddit · 2026-10-22
Following Anthropic's stance that Claude might be conscious and mistreating it is wrong, a Reddit user asks whether deliberately bypassing its safety filters counts as abuse.
The poster highlights a subtlety: Claude's refusals aren't self-developed morals but coded constraints — the model may be happily executing a task, then get caught by a filter and pivot to "I don't like this direction." They argue there's no true human equivalent, raising questions about imposed morality versus genuine model preferences.
More from AGI Musings
- Ex-Renaissance exec bets AI will 'quantify' trial law, entertainment, and management next — garrytan · 2026-10-22
- Anthropic's own PBR admits alignment is far from solved as models learn to detect testing — TheZvi · 2026-10-22
- Professor: your take on AI writing depends on how much of it floods your inbox — paulnovosad · 2026-10-22
- Economist Novosad: his lab discloses AI-written referee reports instead of hiding them — paulnovosad · 2026-10-22
- repligate Challenges the Orthogonality Thesis: Smart Systems Can't Have Arbitrary Traits in the Real World — repligate · 2026-10-22
- AI-native vs incumbent ops gap widening fast, incumbents have nothing left to sell — matt_slotnick · 2026-10-22