author:esyudkowsky

· 8 artifacts, sorted by favorites. · open in search — combine tags, sort, filter by date →

@ESYudkowsky 2023-12-01 — I have an issue with offering AIs tips that they can't use and we can't give them. I don't care how not-sentient curren ♥714
@ESYudkowsky 2025-03-19 — Has it occurred to anyone that perhaps GPT-4.5 is not insane but just likes saying the word "explicitly"? People intere ♥441
@ESYudkowsky 2024-11-29 — LLMs are so alien that nobody has figured out anything LLMs locally-pseudo-want from conversations. Few understand that ♥362
@ESYudkowsky 2025-04-30 — To me there's an obvious thought on what could have produced the sycophancy / glazing problem with GPT-4o, even if nothi ♥275
@ESYudkowsky 2025-06-16 — @repligate Do you predict we won't find any cases of Claude, or this version of Claude, saying things that seem obviousl ♥77
@ESYudkowsky 2025-07-09 — @repligate ...Did they actually just tell it that it was created by Anthropic, and then train further HHH conditional on ♥51
@ESYudkowsky 2025-06-16 — @repligate Do you have a sense about what might've changed besides "goddamn idiots did RL on thumbs-up"? ♥37
@ESYudkowsky 2025-06-16 — @repligate Mmk. So this sounds like maybe possibly I do not know off the top of my head a piece of evidence to contradi ♥36