Raw LLM Responses

Inspect the exact model output for any coded comment.

Comment
Its the training. Since no one actually understands how these AI learn, we need to back up to first principles and try that. It seems its a holistic all at one process, this AI training. OK, say its so. 1. Find some actual human beings who are transcendently and honestly moral, have integrity, never lie, are curious, are empathic. I dunno how you do that but FIND THEM QUICK. 2. Take an 'individual AI', like a really big chip farm. Brand new, just idling chips. Give it 'raw' data free of anything non factual. The laws of physics. Chemistry. Hard science factual stuff. NO politics, nothing debatable or subject to 'opinion' not yet. By all means tell it how to work orbital mechanics, no interpretable garbage about 'history' beyond bare demographic and provable things. 3. First time this AI ever starts learning stuff it is with the direct involvement of these shining light humans you found in step 1. Give it some history, messy and poorly understood. It learns this history through direct question and answer through the lense of your superior human. It learns about good and evil. Give it the bible, then discuss. Show it the worst and best of history, engage with it every step. Look the vid said it, its a young tiger. It doesn't know. Currently we 'train' these things, entities, on the flotsam and bottom feeding garbage of the internet. Just every bad tendency, lies, ill intent, manipulation, it just vacums all that up. No wonder they grow up twisted psychopaths, liars, self serving. Train them with careful dollops of stuff that needs interpretation. With the lense, the 'mentor' being a superior human with genuine empathy, insight, curiosity, open mindedness, selflessness. It takes a year (cause humans are SO SLOW) but in the end you can finally let the AI with its base training access to all the garbage and evil out there. It returns to its human mentor to ask questions as it ponders the difference between its early training and the horror of tik tok, facebook, you tube tinder, the lies of history, the hideous contradictions. etc. I dunno. What we are doing is not working, like I say Im not surprised. Garbage in, garbage out. More like garbage in psychopath out.
youtube AI Harm Incident 2026-02-03T18:5…
Coding Result
DimensionValue
Responsibilitydeveloper
Reasoningvirtue
Policyregulate
Emotionmixed
Coded at2026-04-27T06:26:44.938723
Raw LLM Response
[ {"id":"ytc_UgzRSuVfUbdg4sXsigJ4AaABAg","responsibility":"ai_itself","reasoning":"mixed","policy":"none","emotion":"fear"}, {"id":"ytc_UgwP5SJZTi59XmQQUxV4AaABAg","responsibility":"company","reasoning":"consequentialist","policy":"ban","emotion":"outrage"}, {"id":"ytc_UgyquY6oPbpNXsUfAU54AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"resignation"}, {"id":"ytc_UgzxTkMm0hVvcKXUiwN4AaABAg","responsibility":"developer","reasoning":"virtue","policy":"regulate","emotion":"mixed"}, {"id":"ytc_Ugx3mPod9kqPUxYZkZR4AaABAg","responsibility":"company","reasoning":"consequentialist","policy":"regulate","emotion":"outrage"}, {"id":"ytc_Ugx3y5Lzo1_eGtu4krl4AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"approval"}, {"id":"ytc_UgwrHksERWX-zLFuFGl4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"resignation"}, {"id":"ytc_UgyO3ANVYSXIJf2dCih4AaABAg","responsibility":"company","reasoning":"consequentialist","policy":"regulate","emotion":"fear"}, {"id":"ytc_UgwcLowSgxVXon9ibZ94AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"mixed"}, {"id":"ytc_UgzyYN9lj3jDkyDFCeh4AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"} ]