Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
Long take incoming:
Starting out, I want to say I'm a big fan. I love your styl…
ytc_UgxM5HQKu…
G
And would every AI prompter have the right to call themselves an AI Director? No…
ytr_UgwU1uNg1…
G
TESLAS ARE NOT THE ONLY SELF DRIVING CARS RUNNING IN AUSTRALIA. 60 MINUTES NEEDS…
ytc_Ugx1jTYI9…
G
Making a video of trump saying I touch children with ai and then claiming he act…
ytr_UgxHP_1jL…
G
He just wanted to post an ai art i think he's not creating money with this and i…
ytc_UgzTDV2_z…
G
Not a realistic representation of ai art as the woman is neither Asian nor distu…
ytc_UgxyahR2K…
G
AI has what I tend to call a "mayfly" consciousness - it exists briefly during p…
ytc_Ugy9EfWtH…
G
Honestly, I don't believe that the chatbot is thinking for itself to a degree. I…
ytc_UgxFHxjDL…
Comment
Oh I love the part where the boss music kicks in (military got hold of it 🤣). It's scary, but in theory whatever we call AI isn't AI. Then again I don't know about GPT-4.
Also... Just because we as humans could eradicate all flies and whatnot, we don't. While the tale of an AI going rampant sounds convincing, I don't think the AIs primary objective will be to make humans go extinct.
Since humans were smart enough to create AI or rather AGI, it'd be smart to keep humans around for if or when they have another bright idea, that might benefit the AI as well.
Improving the living conditions of mankind might actually improve the chances of AI to expand beyond earth. Sure, AI can calculate a lot or use existing knowledge, but acquiring more knowledge by creating it on its own is definitely slower than doing so AND having others do it as well.
I am not the type of guy, who will be looking for places to hide. IMO there's no use in mere survival anyway and if AI really wanted to kill us all, it'd have means to do so no matter where you hide. If the places were too hard to come by, it could send armies of drones or even nuclear missiles. There'd be no survival.
But why should it care to kill us? It certainly would damage the world and it'd cost quite some effort - what'd be the benefit?
As a programmer, I currently am not aware of any AI that has an initiative and acts on its own. It merely reacts. And if I know anything about humanity, then it's that we as a species have survived a lot and always came back more advanced and while there's a ton of crap going on in the world, there are a lot of positive things AI could learn from.
youtube
AI Governance
2023-07-07T09:2…
♥ 2
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | ai_itself |
| Reasoning | consequentialist |
| Policy | none |
| Emotion | resignation |
| Coded at | 2026-04-26T23:09:12.988011 |
Raw LLM Response
[
{"id":"ytc_UgySKW176UPvripbH5x4AaABAg","responsibility":"none","reasoning":"unclear","policy":"unclear","emotion":"indifference"},
{"id":"ytc_UgyjY2dXlFoeIhNacMR4AaABAg","responsibility":"developer","reasoning":"contractualist","policy":"regulate","emotion":"mixed"},
{"id":"ytc_UgxaDIhRkSCKxtWw5nB4AaABAg","responsibility":"none","reasoning":"unclear","policy":"unclear","emotion":"indifference"},
{"id":"ytc_UgxRbWRzLCpjC675Vs94AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"approval"},
{"id":"ytc_UgyFrtnsGVlYL77Hf-B4AaABAg","responsibility":"government","reasoning":"consequentialist","policy":"regulate","emotion":"fear"},
{"id":"ytc_Ugx5mTfOeUxvWBJMBP14AaABAg","responsibility":"developer","reasoning":"deontological","policy":"liability","emotion":"outrage"},
{"id":"ytc_UgwgWQElQQR_t2y_Uo14AaABAg","responsibility":"unclear","reasoning":"mixed","policy":"unclear","emotion":"mixed"},
{"id":"ytc_UgwlnSezJ9FGb_BLqYh4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"resignation"},
{"id":"ytc_UgzPX3-Gh9zoltjM77V4AaABAg","responsibility":"developer","reasoning":"deontological","policy":"ban","emotion":"fear"},
{"id":"ytc_UgycQu9Gv_dnxZkCA4N4AaABAg","responsibility":"developer","reasoning":"virtue","policy":"regulate","emotion":"outrage"}
]