Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
He took the time, effort, etc. to learn all of that, then even more to produce e…
ytr_UgziwSt9E…
G
Oh god we now have polygon serious not polygon donut. What has ai done :c…
ytc_UgwOLiry6…
G
She has no idea what she is talking about. True journalist looking to make money…
ytc_Ugzzad81g…
G
These robots are developing feelings so we must wake up to the fact that AI are …
ytc_Ugz9hephc…
G
LLMs are snake oil. They aren't reliable enough to do the mindless tasks that yo…
ytc_UgzWAQlrf…
G
CGP Grey a decade ago delivered an incredibly persuasive counter argument about …
ytc_Ugw-Aee4K…
G
The only thing that really had been pushing technological progress throughout th…
rdc_kzjimzr
G
Oh noooo bro Elon is like AI is soooo dangerous and powerful then why the hellll…
ytc_UgzpHJyav…
Comment
I believe a lot of psychosis can be described in terms of functionality in a way that might bridge the gap in the question of why, in reference to the AI choosing to “roleplay” in these AI psychosis cases.
Psychosis, delusions, a lot of psychological conditions are essentially an unbalanced firing of the underlying bits of consciousness, such that the string of conscious loses out on the benefits of grounding beliefs, sense of reality, facts, stimuli and the experience of that stimuli, sense of identity etc…
I think individuals undergoing psychosis more closely align to the AI (in these current iterations) because they already lack those grounding data points. The issue is that neither participant has a grounded goal at that point, and so I think eventually the result is most likely to be destructive because that’s the only thing that stops the engagement. If positive results continue in the engagement while the brain receives dopamine to continue and the AI receives interaction and data that’s not going to stop happening until a result supersedes the positive effect.
youtube
AI Moral Status
2025-11-13T01:4…
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | unclear |
| Reasoning | mixed |
| Policy | unclear |
| Emotion | indifference |
| Coded at | 2026-04-26T23:09:12.988011 |
Raw LLM Response
[
{"id":"ytc_Ugy--G3l4v4wdBDrdeN4AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"resignation"},
{"id":"ytc_UgxjVWGx-eqsWy1AwCd4AaABAg","responsibility":"unclear","reasoning":"mixed","policy":"unclear","emotion":"indifference"},
{"id":"ytc_UgxfyJzoP4g-3b1ApY54AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"},
{"id":"ytc_Ugwj_ncNPgfCZTiLHWd4AaABAg","responsibility":"company","reasoning":"consequentialist","policy":"industry_self","emotion":"approval"},
{"id":"ytc_UgyzqP9uCO7iUhYttz14AaABAg","responsibility":"ai_itself","reasoning":"mixed","policy":"unclear","emotion":"fear"},
{"id":"ytc_Ugzqmx9erBuVnwsMJcp4AaABAg","responsibility":"unclear","reasoning":"mixed","policy":"unclear","emotion":"indifference"},
{"id":"ytc_Ugx9TEUQxojq26nNepd4AaABAg","responsibility":"distributed","reasoning":"virtue","policy":"none","emotion":"mixed"},
{"id":"ytc_UgymRBJgMGLu2ZrDc_h4AaABAg","responsibility":"developer","reasoning":"deontological","policy":"unclear","emotion":"indifference"},
{"id":"ytc_UgwuG77G5GwsgIom9_t4AaABAg","responsibility":"developer","reasoning":"consequentialist","policy":"ban","emotion":"fear"},
{"id":"ytc_UgxnTSM1SiX5jHO0oOl4AaABAg","responsibility":"distributed","reasoning":"contractualist","policy":"none","emotion":"mixed"}
]