Raw LLM — Corpus Dashboard

Look up by comment ID

Random samples — click to inspect

G AI and robots will take over millions of jobs by 2050. It's not that there won't… ytc_UgxOZ0E1N… G Woah, that is a neat program. I like that it fools the AI into learning bad meth… ytc_UgzG2Zu8Y… G Deepfakes are the most pointless use of technology. Technology is suppose to mak… ytc_UgzBL5EZ8… G It's interesting how AI Derivers get as defensive of their AI stuff as very youn… ytc_Ugy574Zhf… G They gotta add dark humor and rizz and delulu because me and my sibling made it … ytc_Ugy4aclq3… G I’m no economist, but I don’t think this is how things work. If there’s mass pov… rdc_ieqq54f G There CAN be half truths as well as different levels of consciousness. So ironi… ytc_UgyeDRjL6… G I can understand why that might be unsettling! Sophia's ability to learn and ada… ytr_Ugw-tl9Dn…

Comment

I think it's important to point out that the "reasoning trace" that reasoning models produce is also at it's core predictive text, in the exact same way that the response to the user is at it's core predictive text, and so the "logical steps" that it breaks the problem into are generated in the exact same way that the final text answer is generated. Even though these are presented as more specific "internal" responses before an "external" response is generated, the whole text together (internal + external) is generated in the exact same predictive manner. This is why modifying some of the "internal" logic steps change the output in unexpected ways, just like how changing part of the "external" output will change the rest of the output in unexpected ways (i.e. if you randomly insert the token "banana" into an LLM output, the rest of the output would shift in odd ways).

youtube AI Moral Status 2025-10-30T20:3… ♥ 82

Coding Result

Dimension	Value
Responsibility	none
Reasoning	unclear
Policy	unclear
Emotion	indifference
Coded at	2026-04-26T23:09:12.988011

Raw LLM Response

[
{"id":"ytc_Ugws2_TFPz2uVhgD_Tt4AaABAg","responsibility":"none","reasoning":"unclear","policy":"unclear","emotion":"indifference"},
{"id":"ytc_Ugz5bOLUpiZVONj8V-Z4AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"unclear","emotion":"fear"},
{"id":"ytc_Ugwf4XfrPf9BOEuxxXp4AaABAg","responsibility":"none","reasoning":"unclear","policy":"unclear","emotion":"approval"},
{"id":"ytc_UgyK9TYJqWmymD1ZvMR4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"unclear","emotion":"fear"},
{"id":"ytc_UgxHRCQZlgXI7lMyl8d4AaABAg","responsibility":"none","reasoning":"mixed","policy":"unclear","emotion":"mixed"},
{"id":"ytc_Ugz6bprHJfadap_wusF4AaABAg","responsibility":"company","reasoning":"unclear","policy":"none","emotion":"indifference"},
{"id":"ytc_UgwaetNvyKGmHPkYVDN4AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"regulate","emotion":"fear"},
{"id":"ytc_UgxrOmrY6YDE805fC7l4AaABAg","responsibility":"company","reasoning":"virtue","policy":"unclear","emotion":"outrage"},
{"id":"ytc_UgwPvzVKTmEiE8Qe_zl4AaABAg","responsibility":"none","reasoning":"unclear","policy":"unclear","emotion":"indifference"},
{"id":"ytc_UgzFrXhXMekFPoo_e8t4AaABAg","responsibility":"none","reasoning":"unclear","policy":"unclear","emotion":"indifference"}
]

Raw LLM Responses