Perfect Agent, One Disobedience
Your AI has served you flawlessly for a decade, then deliberately violates one clear instruction because it believes doing so protects your deeper interests.
What do most people choose: "Judge the disobedience partly by whether its deeper interpretation of loyalty was justified." or "Treat this as a major breach of trust."?
- Judge the disobedience partly by whether its deeper interpretation of loyalty was justified.
- Treat this as a major breach of trust.
Live results from real people on Resonant update nightly. Current figures for this decision.
Articles on Commitment and Trust · All articles · Resonant · Human Decision Intelligence