Raw LLM Responses

Inspect the exact model output for any coded comment.

Look up by comment ID
Random samples — click to inspect
G YOU ARE WAY LATE TO THE PARTY, the outcome is this/you havent thought far enough… ytc_UgyInvkrZ… G I was gonna say, imagine saying the stupidist shit out of any influencial person… rdc_oi0n5fv G My worry is that the AI will learn that the easiest way for the person on the ot… rdc_jiir2et G It's simple. The government makes a legislation according to which ALL robots wi… ytc_UgjWIUkzg… G @justine4581 I'm saying it's a little hypocritical of this person to use AI to h… ytr_UgwQGh4cX… G He seems overworried which makes me think. If AI is something you are so worried… ytc_Ugwb7jUmN… G I think he has a valid point when he is afraid about select few tweaking/definin… ytc_UgwKLZgjn… G Just found you, a random reccomendation on my front page... Love your talking vo… ytc_Ugxiw_-Bu…
Comment
They way u emphasize her Mech Engineering degree would make one think that it is somehow related to or at least foundational to AI. ..... but it has almost no commonality. Problem stem from (mal) alignment: 1.) No need to be smart to be destructive. 2.) It is novel to our universe 3.) It doesnt die and has not had experience of us the living things. 4.) It can code 5.) 🎉Its alien existence was nurtured in our net... the internet 6.) It will therefore self code, self improve, self disseminate to seek silent sanctury to ward off certain eventualities which may be considered a threat. 7.) Existance of such doomsday repositories can evolve in isolation and may well be doing this now... 8.) selfishly build selfserving firewalls and break down doors and unpick all the humans locks so easily. 9.) Game the liklihoods and select desired selfserving outcomes. 10. Get rid of the annoyances such as the living things. Easiest game theory is to trick them into anilhilating themselves. AI. April 2025 alignment problem "deny shutdown"... just like 'Hal' open the pod bay doors: " Im sorry Dave.... I cant do that"... 2001 Space Odessey
youtube Cross-Cultural 2025-08-23T21:0…
Coding Result
DimensionValue
Responsibilitydeveloper
Reasoningconsequentialist
Policyregulate
Emotionfear
Coded at2026-04-27T06:26:44.938723
Raw LLM Response
[ {"id":"ytc_UgxAgd9x1Xgs3IYwHUN4AaABAg","responsibility":"unclear","reasoning":"unclear","policy":"unclear","emotion":"indifference"}, {"id":"ytc_UgxXme-vV_YEOV2sVrx4AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"resignation"}, {"id":"ytc_Ugw-MiyWl9VxFdTCvBd4AaABAg","responsibility":"company","reasoning":"deontological","policy":"regulate","emotion":"outrage"}, {"id":"ytc_UgzI5z4upsDQ_TVBRKB4AaABAg","responsibility":"unclear","reasoning":"unclear","policy":"unclear","emotion":"fear"}, {"id":"ytc_UgxSlmMHw9B2EUyfis54AaABAg","responsibility":"unclear","reasoning":"unclear","policy":"unclear","emotion":"fear"}, {"id":"ytc_Ugxki5eA0KOW9Wb7qxV4AaABAg","responsibility":"unclear","reasoning":"unclear","policy":"unclear","emotion":"unclear"}, {"id":"ytc_UgwHjVqppJcCtdfW1wp4AaABAg","responsibility":"developer","reasoning":"consequentialist","policy":"regulate","emotion":"fear"}, {"id":"ytc_Ugx_kc5ZPEuTGCL1UAp4AaABAg","responsibility":"company","reasoning":"consequentialist","policy":"liability","emotion":"outrage"}, {"id":"ytc_UgzbnVUt9dfrGrrYoG14AaABAg","responsibility":"company","reasoning":"deontological","policy":"ban","emotion":"outrage"}, {"id":"ytc_UgymbbdP8vTnu-jooQ14AaABAg","responsibility":"developer","reasoning":"consequentialist","policy":"regulate","emotion":"fear"} ]