AI ResearchMachine Learning1 min reading time

Character Training for Risk-Averse Agents

LessWrong
Read the full article
Researchers propose a method called Character Training to develop AI agents that exhibit risk-averse behavior, aiming to improve decision-making under uncertainty by aligning agent actions with cautious strategies.

More in AI Research

AI Research4 min read

This new theory on how AI ‘thinks’ could impact whether it kills us all

Business Insider
AI Research4 min read

The New Science Of Machine Attention

Forbes
AI Research10 min read

He Wants AI Benchmarking To Be a Science, Not a Dark Art

Quanta Magazine