We just published a long paper on Truthful AI (overview post). We’ll be running an Ask Me Anything on truthful AI from Tuesday (October 26) to Wednesday (October 27) at this post.
You may wish to ask about:
- Anything discussed in the Truthful AI paper (e.g. concepts, costs/benefits, governance, development, connections to beneficial AI).
- Truthfulness for current language models (e.g. TruthfulQA — which Owain co-authored)
- Promising research directions on truthful AI
- Anything else (unrelated to truthful AI) — although we are more likely to answer questions about truthful AI.
If you want to ask something just post a top-level comment. Also feel free to make comments that are not questions (e.g. objections or suggestions).