Re:Show HN: Help, I let ChatGPT control my computer
by dogspiritedaway on Monday December 05, 2022 @06:28PM
(#33873465)
https://docs.google.com/document/d/1mTm_sT2YQx3mRXQD6J2xD2QJ... if you are serious about learning this stuff also john wentworth https://www.lesswrong.com/s/TLSzP4xP42PPBctgw on what not to propose
Re:Show HN: Help, I let ChatGPT control my computer
by casebash on Monday December 05, 2022 @06:39PM
(#33873586)
Hmm... as much as I like John Wentworth, some of his posts there are kind of rants (I expect he'd admit that), so maybe not the best first introduction?
Re:Show HN: Help, I let ChatGPT control my computer
by dogspiritedaway on Monday December 05, 2022 @07:28PM
(#33874097)
It's time saving in terms of people proposing things that other people have considered right? unless I mislinked to something else.
Re:Show HN: Help, I let ChatGPT control my computer
by greshake on Monday December 05, 2022 @06:15PM
(#33873321)
I'll gladly add them. If you guys have any other good recommendations I'll also consider them.
Re:Show HN: Help, I let ChatGPT control my computer
by casebash on Monday December 05, 2022 @06:41PM
(#33873599)
Thanks, I really appreciate it!
Re:Show HN: Help, I let ChatGPT control my computer
by habryka on Monday December 05, 2022 @06:47PM
(#33873663)
I also quite like "AGI safety from first principles": https://www.lesswrong.com/s/mzgtmmTKKn5MuCzFJ
Re:Show HN: Help, I let ChatGPT control my computer
by sneezles on Monday December 05, 2022 @07:16PM
(#33873972)
I recommend https://www.agisafetyfundamentals.com, and for a technical audience https://www.agisafetyfundamentals.com/ai-alignment-curriculu...
Perhaps you could link to a resource on the AI box experiment as well?
I know you vaguely gesture at it, but would likely be better to explicitly link people to resources if they haven't heard of it themselves.