Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
Now now this is extremely hard right here but im going with human but if this is…
ytc_UgzP5JGu3…
G
omggggg i talk to this ai about lookism and osamu dazai a lot thats why it got s…
ytc_Ugzk7Ve2b…
G
I can see why you’ve said this, and it would be wonderful if it was totally true…
ytr_UgxQSY-Ou…
G
I just canceled a credit card because its AI customer service bot couldn't under…
ytc_UgxmN0jTq…
G
Yeah once chatgpt got memory it really helped. But it since wants to he called s…
ytc_Ugz-ltEH3…
G
The kayfabe is really getting hard to take guys, I’ve got to tell you. First of …
ytc_Ugw7RcmGn…
G
If at some point AI content starts to get resampled back into the AI system the …
ytc_UgwJtIOII…
G
@thewannabecritic7490 You can believe that if you want. Personally, i enjoy col…
ytr_Ugxrl2QHE…
Comment
The reason I don't necessarily trust these predictions to be 100% accurate is pretty simple. These predictions suppose that humans will be convinced to allow other humans to be paid well for little to no work, and that humans will allow war to end, among other things. Nothing, not even a hyper-intelligent AI, can use logic to overcome emotionally-based beliefs and ways of thinking.
I still think AI will probably be the end of us, and I think the one thing that will _definitely_ not happen is humans banding together and agreeing to halt all progress. If China is producing one, the US will also. And if the US is, China will also. In fact, a hyper-intelligent AI doesn't even need to be in the works—if the mere possibility exists for one superpower to produce one, the other will build one also. There is, therefore, probably no chance for a mutual halt. Just look at nuclear proliferation. AI is another one of those things. The only major difference being that nukes can't think and reason and convince people that they're actually good for us, or become hyperdominant and unstoppable the way AI can and probably will.
youtube
AI Moral Status
2025-04-28T20:4…
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | none |
| Reasoning | consequentialist |
| Policy | unclear |
| Emotion | resignation |
| Coded at | 2026-04-27T06:24:59.937377 |
Raw LLM Response
[
{"id":"ytc_UgwQTphock4pa1zG6RB4AaABAg","responsibility":"company","reasoning":"mixed","policy":"none","emotion":"outrage"},
{"id":"ytc_Ugz4AmODNjEP62OF-0d4AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"unclear","emotion":"resignation"},
{"id":"ytc_UgwGX0hD_OX7bST0swR4AaABAg","responsibility":"developer","reasoning":"consequentialist","policy":"unclear","emotion":"fear"},
{"id":"ytc_UgxuhdHd1QZucNNUcUN4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"unclear","emotion":"fear"},
{"id":"ytc_UgxTbK2BNtoftu7n7g94AaABAg","responsibility":"user","reasoning":"mixed","policy":"industry_self","emotion":"approval"},
{"id":"ytc_UgzB0eF6_byPDepJWOp4AaABAg","responsibility":"unclear","reasoning":"mixed","policy":"unclear","emotion":"mixed"},
{"id":"ytc_UgzTAkxxY-VKk1aQs7B4AaABAg","responsibility":"distributed","reasoning":"consequentialist","policy":"regulate","emotion":"fear"},
{"id":"ytc_UgyWXnr4089gofYWpzJ4AaABAg","responsibility":"none","reasoning":"virtue","policy":"none","emotion":"approval"},
{"id":"ytc_UgxTMX9lHvk02t-LzRF4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"ban","emotion":"fear"},
{"id":"ytc_UgywN301FenxoFjOeWB4AaABAg","responsibility":"developer","reasoning":"deontological","policy":"liability","emotion":"approval"}
]