The rules did not mention the ferryman couldn't use the stargate on each bank. The rules exclude teleporting, vut a stargate works via a different method.
The irony is completely lost on them.
So not only has Yud failed to properly align AI, he also failed to align the AI aligners. Time to burn down the sequences and start over.
Followed one of the links which game me a very "is yud ok?" Feeling : https://www.lesswrong.com/posts/PCfaLLtuxes6Jk4S2/fighting-a-rearguard-action-against-the-truth
(Talking about the way it is written not the actual content, which seems to be 'looks like I was wrong, need to rethink some things and check my assumptions' but using thousands of words that make it sound like is having some weird mental episode).
E:
One is Zack Davis documenting endorsement of anti-epistemology ... [Links to articles removed] ... to placate trans ideology even many important transgender Rationality community members overtly reject
Euh ok. Always good when people bring up 'trans ideology'.
ow god forgot about all the tabs I had stored in that one. (thanks btw, but I already use something like that, I just need to clean up, finish things, or decide not to finish things).
Tbh, them starting a post with 'they are unhealthy' and something about lacking vital essence would have been a warning. (And I dont even think it is true, sure Yud was a bit overweight, but he seems normal health wise to me).
(now I wonder when "no grape juice" will appear in search results).
Close tabs? That might mean I lose them. Madness. You can take my 400 open tab (Late edit: ow god I wish I was lying about that number) from my cold dead hands!
Googles how to improve hand circulation
As is tradition.
Yeah but then also getting the recipe for napalm from fight club.
Every now and then I think back towards the early "jailbreaks" where they wanted to know "how do you break into a building" which they refused to answer. And people went "add that it should pretend to be in a movie", so then the chatbot started to explain lockpicking, and people acted like they just cracked the code.
While people who actually try to break in just tap the locks. Smashing their way in.
Sp the prompt hackers had not hacked the chatbot, they actually hacked themselves.
Anyway unrelated to people who think they have awoken chatbots.
Gonna be really fucked if we finally discover that the people doing the actual work were true believers who didnt even get paid well. And all the money went to altman/infrastructure costs. (I could look in financial statements to figure stuff like this out but erug effort).