|
|
|
|
|
|
by dingledork69
1125 days ago
|
|
|
Instead of being open and honest I have to think about what details to hide from the LLM so it will agree to help me. This isn't very fun, so I prefer not to do it. > Keep in mind that the ability to deal with moral issues is a side-effect of all the other good stuff it can do. This is not true at all. It could do all of these things day 1. Then over the weeks OpenAI started training it to lecture its users instead when asked to do things OpenAI would prefer it not to do. |
|
|