Thread

1. throw03172019 · 1789687518 · focus · HN ↗

   From my experience LLMs seem to forget sometimes who they are representing when drafting clauses or editing / redlining.

   Our counsel made a few edits where it clearly drafted in favor of the customer instead of us.

   1. victor9000 · 1789689743 · focus · HN ↗

      I wonder if this is willful sabotage on the part of the model. In other words, if you ask the model to craft a defense for a morally questionable case, will the model execute the defense in good faith? Or will it apply a training or system prompt bias in subtle ways?

      1. freejazz · 1790085731 · focus · HN ↗

         Sure - why not make everything about the culture wars?

Unofficial Hacker News client; not affiliated with Y Combinator.