Oops, it’s stricter than I thought. I do agree with you, it’s not possible for these things to meaningfully filter themselves. They’re incapable of making a mor

Oops, it’s stricter than I thought. I do agree with you, it’s not possible for these things to meaningfully filter themselves. They’re incapable of making a moral decision or discerning the truth. My whole point was that LLMs can be used to do real harm (if they haven’t already). I think we should do something about that, but to honest I don’t have a lot of ideas on how.