AIBullisharXiv – CS AI · Mar 177/10
🧠
Emotional Cost Functions for AI Safety: Teaching Agents to Feel the Weight of Irreversible Consequences
Researchers propose Emotional Cost Functions, a new AI safety framework that teaches agents to develop qualitative suffering states rather than numerical penalties to learn from mistakes. The system uses narrative representations of irreversible consequences that reshape agent character, showing 90-100% accuracy in decision-making compared to 90% over-refusal rates in numerical baselines.