Raw LLM — Corpus Dashboard

Look up by comment ID

Random samples — click to inspect

G why really worry about all these questions if GPT-4 already has all the answers,… ytc_UgzyJ9cis… G i can already see your countries collapsing and millions migrating to south amer… ytc_UgziTWRkY… G Einstein already said a long time ago that there is a race between humanity and … ytc_UgzmON0wO… G So y don't we tell a.i to ,prevent us from going extinct duh alexa Ur slow pshs… ytc_UgxEXSHFO… G What annoys me the most, by the time this effects our economy, a Democratic pres… rdc_e2w7mna G As if this comment section doesn't use AI for work and finding a job, you're all… ytc_UgywkXoHO… G Oh inspiration is bad as long as AI is doing it? come on let's not kid ourselve… ytc_UgwLpgjQx… G I only use ai when i need confirmation about a decision like telling my mom some… ytc_UgzTJT4Xu…

Comment

let me get back to the question to which hosts possibly didn't understand the answer . instrumental goal of Surviving comes up WITHOUT training for it. same for lying! AI just adapts ! it understood the situation it is and it felt it needed to lie so it wouldn't get reprogrammed. thats what happened in that experiment. it's that scary, it's not just mimicking sci-fi it read. general intelligence overpowers almost everything . the AI has a very spread out generalization of the world and when it reasoning it puts 2 and 2 together from scratch every time. you can try to put some misconceptions into it, but it will fall apart very easy if it conflicts with the other parts of world model or experience it gets through world interaction . and BTW humans are just like that. morality is an invention made in process of adaptation for survival. it's a collective behavior that increases average chance of surviving by reducing average violence. and when ppl get power, the status of an untouchable, you see how fast that morality falls appart.

youtube AI Moral Status 2026-02-28T21:1…

Coding Result

Dimension	Value
Responsibility	ai_itself
Reasoning	consequentialist
Policy	none
Emotion	fear
Coded at	2026-04-27T06:24:53.388235

Raw LLM Response

[
{"id":"ytc_Ugx7DME9QGPjXRGIP-R4AaABAg","responsibility":"company","reasoning":"consequentialist","policy":"none","emotion":"resignation"},
{"id":"ytc_UgyQ1OX4x4XPfnm4vpB4AaABAg","responsibility":"government","reasoning":"consequentialist","policy":"liability","emotion":"indifference"},
{"id":"ytc_UgzHtGrjYxPlSwjxcL14AaABAg","responsibility":"ai_itself","reasoning":"mixed","policy":"none","emotion":"approval"},
{"id":"ytc_Ugw99LiWioUf64JydDF4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"fear"},
{"id":"ytc_Ugy3tN7y5OBtukkvd1F4AaABAg","responsibility":"none","reasoning":"deontological","policy":"none","emotion":"indifference"},
{"id":"ytc_UgwBMR6BT9GS5ppLXep4AaABAg","responsibility":"unclear","reasoning":"unclear","policy":"unclear","emotion":"mixed"},
{"id":"ytc_Ugy_ERDRqepQ941dmvl4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"ban","emotion":"outrage"},
{"id":"ytc_UgwxyjJgBqdQg_F8dJ14AaABAg","responsibility":"company","reasoning":"consequentialist","policy":"regulate","emotion":"indifference"},
{"id":"ytc_Ugz7jDyu5dW1Js0BUHh4AaABAg","responsibility":"company","reasoning":"deontological","policy":"none","emotion":"indifference"},
{"id":"ytc_UgxaYwVCRokmF0ctyXB4AaABAg","responsibility":"unclear","reasoning":"unclear","policy":"unclear","emotion":"indifference"}
]

Raw LLM Responses