Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
I’ll believe this would definitely happen under Trump presidency when a presiden…
ytc_Ugx7vnNvu…
G
@Chicago48 U mean, STEAL images & designs by copying thm as resource material si…
ytr_UgwJUuAwb…
G
Im currently working on my memoir and i used Ai to sharpen & polishing my word a…
ytc_Ugw9lEQW3…
G
50:25 "I don't see how it can get agency". She really needs to understand…
ytc_UgyJKAzv2…
G
Neither.
We'll enter into a fragmented age of despots and corporatism where goo…
rdc_dulen2m
G
The obvious true danger is military AI that are constantly evolving and being in…
ytc_UgyiJZll2…
G
Ai can do a lot which is cool i can cheat on my tests and school work, but its n…
ytc_Ugxb4cFkp…
G
We appreciate your observation! Sophia's ongoing development includes refining h…
ytr_UgzzijlNS…
Comment
Just to be clear here, you can't take away from that paper that "models store copies." Only very specific books were tested, because they are over represented on the internet already, books like Harry Potter and 1984. You would not be able to generalize the paper's jailbreaking process to extract other books because most books will not be overrepresented in the training data.
This is overfitting, and is undesirable even for a purely self-interested AI company because overfitting degrades model performance. A model wasting capacity memorizing Harry Potter verbatim is learning fewer generalizable patterns. The fix is straightforward data curation, better deduplication, preventing over-representation of any single work. The paper shows some companies were sloppy about this, but no generalizable conclusion relating to learning from copyrighted material.
youtube
2026-01-23T23:1…
♥ 1
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | none |
| Reasoning | mixed |
| Policy | unclear |
| Emotion | indifference |
| Coded at | 2026-04-27T06:26:44.938723 |
Raw LLM Response
[
{"id":"ytr_UgwxcKl4B92AvDU3GJp4AaABAg.AS3GKO2L0LZAS5-6AgsPRX","responsibility":"none","reasoning":"unclear","policy":"unclear","emotion":"indifference"},
{"id":"ytr_UgwxcKl4B92AvDU3GJp4AaABAg.AS3GKO2L0LZAS7DZGyXZL0","responsibility":"government","reasoning":"deontological","policy":"liability","emotion":"outrage"},
{"id":"ytr_UgwTxn5nWOeZSciYSbx4AaABAg.AS342zgtgCSASLGP6uKRul","responsibility":"none","reasoning":"mixed","policy":"unclear","emotion":"indifference"},
{"id":"ytr_UgwTxn5nWOeZSciYSbx4AaABAg.AS342zgtgCSASLbh98lduO","responsibility":"company","reasoning":"consequentialist","policy":"regulate","emotion":"fear"},
{"id":"ytr_UgwTxn5nWOeZSciYSbx4AaABAg.AS342zgtgCSASNFZGnJw0j","responsibility":"company","reasoning":"mixed","policy":"industry_self","emotion":"approval"},
{"id":"ytr_UgwYMh4HWSnqiJIvuZ54AaABAg.AS2ytOWq-jJAS6b6C9mzQk","responsibility":"ai_itself","reasoning":"deontological","policy":"none","emotion":"approval"},
{"id":"ytr_UgxTpwOqlCLgISlJwXZ4AaABAg.AS2wksN_TYgAS3haijVLxT","responsibility":"company","reasoning":"deontological","policy":"liability","emotion":"outrage"},
{"id":"ytr_UgxTpwOqlCLgISlJwXZ4AaABAg.AS2wksN_TYgAS6dkEe4Tvm","responsibility":"none","reasoning":"mixed","policy":"none","emotion":"indifference"},
{"id":"ytr_UgyuLBQvV1IdDSyxP1F4AaABAg.AS2vzUWJEb_AS6ejj7VD7d","responsibility":"none","reasoning":"mixed","policy":"unclear","emotion":"resignation"},
{"id":"ytr_UgwyahGFIVuQXcs44aJ4AaABAg.AS2tE9KfDXEAS2tnpYQJw7","responsibility":"company","reasoning":"unclear","policy":"ban","emotion":"outrage"}
]