This post presents a further development of the reduced impact AI approach, bringing in some novel ideas and setups that allow us to accomplish more. It still i
New AI safety ideas.
See on lesswrong.com
Louie Helm is a Machine Learning Engineer