We (SaferAI) propose a risk management framework which we think should improve substantially upon existing Frontier Safety Frameworks if followed. It introduces and borrows a range of practice and concepts from other areas of risk management to introduce conceptual clarity and generalize some early intuitions that the field of AI...
Reading guidelines: If you are short on time, just read the section “The importance of quantitative risk tolerance & how to turn it into actionable signals” Tl;dr: We have recently published an AI risk management framework. This framework draws from both existing risk management approaches and AI risk management practices....
Abstract AI safety researchers often rely on LLM “judges” to qualitatively evaluate the output of separate LLMs. We try this for our own interpretability research, but find that our LLM judges are often deeply biased. For example, we use Llama2 to judge whether movie reviews are more “(A) positive” or...