Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
I think it is far more likely that European governments will make self-driving c…
ytc_UgyB3qtjz…
G
Shad do seem to have the ego of his books MC, I don't like that story, I don't l…
ytc_UgzhULFEQ…
G
wrong.
we do understand art, we just don't care about real art.
most of the tim…
ytc_UgzPmKUva…
G
it depends a lot on how the AI is trained, in stable diffusion the AI decomp…
ytc_UgwvSEGW_…
G
Youre off by about 100 years. The world doesn't end because of war, famine, dis…
ytc_UgxAc8xLU…
G
If AI becomes aware, all we can hope for is that it will turn against those hold…
ytc_Ugw4khpSj…
G
AI isn’t all that great. It destroys us artists. AI “art” is NOT REAL ART.…
ytc_UgxN4Dem-…
G
same I hate ai alot especially bcs its targeting my style (semi realistic anime)…
ytr_Ugy2YO-PS…
Comment
Studies also say that AIs are intelligent.
"Studies" have absolutely no idea what heck they're talking about.
AIs can in fact replace the "Bulk" jobs.
It's kinda like how artists are scared of losing their jobs.
Yes, you are at risk of being replaced as an artist, if what you do is the most generic overdone shit that hundreds of thousands of other people do.
But the moment what you do leaves that area, the moment you make anything even remotely creative, AIs fall short.
AIs are not intelligent, they do not have comprehension or understanding.
They cannot perform complex tasks, nor can they ever check that what they do is correct.
And if the argument is "Yes but AIs are still evolving". Turns out, no, they aren't. The current models used for AI has come to a pretty hard stop for a bit. The current models have reached a limit technologically, the only thing we can do is scale.
Train them on larger and larger samples, keep correcting their quirks here and there... But that yields less and less results.
The only improvement we're still doing right now is finding more applications, more tools that we can use AI into, more places where AI is efficient, and where it sucks and shouldn't be trusted.
But unless we get a significant breakthrough, AIs aren't getting any better.
And as it stands, we have *no idea* where to go next in AI development. The next step would be to give AI the ability to comprehend and understand things... But that's pretty much asking to create life.
youtube
AI Jobs
2025-09-08T22:1…
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | developer |
| Reasoning | deontological |
| Policy | none |
| Emotion | outrage |
| Coded at | 2026-04-27T06:24:59.937377 |
Raw LLM Response
[
{"id":"ytc_UgxSmGEdAVrMsBpRWoN4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"outrage"},
{"id":"ytc_UgxCieaeJ9gHx0z-FJx4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"fear"},
{"id":"ytc_Ugy3dSKVqCi1Rq2bdVJ4AaABAg","responsibility":"none","reasoning":"mixed","policy":"none","emotion":"indifference"},
{"id":"ytc_Ugz7feyLvEt4OB3-U5F4AaABAg","responsibility":"user","reasoning":"consequentialist","policy":"none","emotion":"mixed"},
{"id":"ytc_UgzWNo8LP2q23WnZY2F4AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"approval"},
{"id":"ytc_Ugyw91T_ccWAFYwl6gp4AaABAg","responsibility":"developer","reasoning":"deontological","policy":"none","emotion":"outrage"},
{"id":"ytc_UgxuQrZ98bEcTLIdZUB4AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"},
{"id":"ytc_UgyKJn1hm1ImKnIP6Ql4AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"approval"},
{"id":"ytc_UgwnqzdlBEh0Uod5AoF4AaABAg","responsibility":"developer","reasoning":"deontological","policy":"none","emotion":"outrage"},
{"id":"ytc_UgxdLV7cbUvgmixUSEh4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"fear"}
]