Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
You fools! You've just posted more food online for the AI machine to gobble up -…
ytc_UgwR6fpQd…
G
AI is different, if not regulated, AI can crack all the passwords in all accoun…
ytr_UgyfpA7A5…
G
Yall don’t seriously think that giving AI those parameters is going to reveal Th…
ytc_Ugz1OAl5c…
G
Yeah, it is not about AI, is about humans and what they will do with it. We are …
ytc_UgwJG9WvS…
G
@dedhampster4730 Another problem is that if I do make AI-generated images and wa…
ytr_Ugzs3GD6n…
G
It means current AI owners are all vying to be the next "Rockerfeller" and like …
ytc_UgwhDVyR2…
G
@gierdziui9003 this is different, milling is a mechanical and repetitive process…
ytr_Ugyu7bJBp…
G
@GorgieClarissaI’m saying that mistakes happen. More often from humans than not.…
ytr_UgwjPS_x-…
Comment
I mean, honestly it looks like an AI based filter to "improve" image quality. It's not quite generative AI in the sense of a deep fake. It isn't adding fingers or creating things that just don't make sense. To me it doesn't have all the same hallmarks of a deepfake, just the slight uncanny valley effect. It's just warping the base content, much like a filter. My guess, which I am fairly confident on, is that YouTube is downscaling the videos, possibly also cutting frames, and then using AI based tools to upscale or add the frames back in. Why? Storage space. We all think storage space is cheap until it is used on a scale like YouTube. Video games have been using similar technology for a while now. I think that legally YouTube would not have to disclose it since it isn't fundamentally altering the content or creating fake content. That's not a defense of it. I don't think it is remotely as bad as, for example, Spotify releasing their own AI generated music, but I am still not a fan and don't think it is a good idea. Also, consider that companies make decisions that don't actually benefit them logistically all the time. I know some people like to say things like "they wouldn't be doing it if they didn't run the numbers and it was beneficial" but that ignores the fact that not all purely numerical decisions are a good idea.
youtube
2026-01-08T17:3…
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | unclear |
| Reasoning | mixed |
| Policy | unclear |
| Emotion | indifference |
| Coded at | 2026-04-26T19:39:26.816318 |
Raw LLM Response
[
{"id":"ytc_UgyefQunjyl5Cl6pzWl4AaABAg","responsibility":"company","reasoning":"consequentialist","policy":"liability","emotion":"outrage"},
{"id":"ytc_UgyQdzeE6ICzTYVTcpN4AaABAg","responsibility":"none","reasoning":"unclear","policy":"regulate","emotion":"outrage"},
{"id":"ytc_UgytIf5aPoEh9T3OAfN4AaABAg","responsibility":"unclear","reasoning":"mixed","policy":"unclear","emotion":"indifference"},
{"id":"ytc_UgyFxV_W9E0OyFDSdst4AaABAg","responsibility":"company","reasoning":"consequentialist","policy":"none","emotion":"fear"},
{"id":"ytc_UgxDTYCsLweaB_YvvS94AaABAg","responsibility":"company","reasoning":"deontological","policy":"liability","emotion":"outrage"}
]