This is an update on the work on AI Safety via Debate that we previously wrote about here. …| www.alignmentforum.org