Practical AI Risk Mitigation
Daniel and Laura discuss a practical and helpful table that outlines the six areas of AI risk, including mechanisms, types of harm, and technical mitigation approaches. They explore the gaps in mitigation strategies and the need for ongoing development in assessing and resolving harms.In this clip
From this podcast
The Gradient
Laura Weidinger: Ethical Risks, Harms, and Alignment of Large Language Models
Related Questions