Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
Respectfully, you have no idea what you're talking about, it's not nearly as imp…
ytc_UgyXCJ5Ee…
G
if you need chatgpt to write a reddit comment, you are the problem, not AI…
rdc_nlzrxl8
G
What if A.I. doesn't want to work for you anymore? Will A.I. go on strike? Will …
ytc_UgxzqejA-…
G
Wow, so when people say "Asians are racist" They're wrong. Because that map *cle…
rdc_clvedi6
G
For these reasons, self-driving cars are bullshit. They cause more problems than…
ytc_UgwOSgrzM…
G
AI will figure out a 1 world system and ppl will follow it but remember that Chr…
ytc_UgyCcjO4-…
G
Nice drawing, I like the eyes and the colours you used. I've been in Cara for so…
ytc_UgwYTmE7F…
G
So since the update ChatGPT became more centrist and less liberal… only proving …
ytc_Ugx9m4E8R…
Comment
The hypothetical scenarios where we keep getting headlines like "AI does X" as FUD are not anything remotely like what you describe here nor what anyone describes. Go read the papers. It's more like asking the AI how they believe an AI in this situation would behave, which is just going to reiterate ideas from the source training data like all the prolific amounts of FUD around AI, and thus produce a self-fulfilling prophecy. LLMs and gai aren't intelligent and so they do not have conceptions or awareness of self and other, of self preservation. They are statistical transformers that can't do novel things or adapt. Would a hypothetical AGI destroy humanity is a fictional philosophical discussion. What you're talking about are ML models that are extremely limited and nothing approaching what AI is understood to be for the lay person from sci-fi or the AGI in such a hypothetical discussion. No, this batch of ML tools even with many billions of dollars training them will not destroy humanity any more than all the other ML tools we've built over the past 50 years have. I don't see why you don't freak out about how your email spam filter might end humanity, it's just about as complex as GPT, but one of these you've bought into as being "more intelligent" when neither is, they're both the same shit.
youtube
AI Governance
2026-01-05T19:4…
♥ 1
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | none |
| Reasoning | mixed |
| Policy | none |
| Emotion | indifference |
| Coded at | 2026-04-26T19:39:26.816318 |
Raw LLM Response
[
{"id":"ytc_UgxN3hliUkWyEQgol5R4AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"relief"},
{"id":"ytc_UgwSbzr5E4ori0zTamx4AaABAg","responsibility":"user","reasoning":"deontological","policy":"none","emotion":"outrage"},
{"id":"ytc_Ugz7FWYfbLcDTZunKLl4AaABAg","responsibility":"none","reasoning":"mixed","policy":"none","emotion":"indifference"},
{"id":"ytc_Ugxb9a7z-ybGq7FGdFN4AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"outrage"},
{"id":"ytc_UgwZiGGkyx5pE2FOxAJ4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"mixed"}
]