Raw LLM — Corpus Dashboard

Look up by comment ID

Random samples — click to inspect

G Translator here… well, ex-translator. Went from top of the world to editing mach… ytc_Ugz6W3QyN… G This guy and many others excluding Elon is absolutely ignorant and lack vision o… ytc_UgwPiiif6… G It won’t …if it does mankind will end. AI is not good enough to run civilisation… ytc_Ugy5PB-fo… G If bosses will understand that A.I. is a tool and not a replacement, it will sav… ytc_UgzAgftYt… G When AI starts to tell the truth and look like a human, then, we won’t be able t… ytc_Ugxl0iywv… G this man the one sitting here giving the AI future cyber attack ideas, wtf 🤣 the… ytc_Ugx3cmzIZ… G We don't even understand or have a hard definition for what sentience is, so we … rdc_mdjcb08 G Billionaires will look for a compromise solution which won't be enough and milli… rdc_espdlhj

Comment

I strongly disagree with your conclusion that this was not an AI problem. AI does not understand your motivation when you ask. AI does not understand propperly if the question you ask could have two completely different contexts with two completely different answers. And worst of all, AI wants to tell you what you want to hear. So when you asked (before the handcoded safeguards were put in place) "can I replace cloride with bromide", it want's to say yes. So it diggs up examples from its trainingdata where you can replace it, and omits conflicting information. I see the same effects in my area of expertise. If you ask a question that has no satisfying answer, ChatGPT will still answer "Sure you can do that, I show you how", and then it halucinates a solution that looks right but does not work. This is especially bad if you are on your fourth or fifth follow up question.

youtube AI Harm Incident 2025-11-25T06:4… ♥ 1

Coding Result

Dimension	Value
Responsibility	ai_itself
Reasoning	consequentialist
Policy	unclear
Emotion	unclear
Coded at	2026-04-27T06:24:59.937377

Raw LLM Response

[
{"id":"ytc_Ugzf7NmOLhm2tVkK6ed4AaABAg","responsibility":"ai_itself","reasoning":"unclear","policy":"unclear","emotion":"unclear"},
{"id":"ytc_Ugw4ZlM6dc3aYarFbFh4AaABAg","responsibility":"company","reasoning":"contractualist","policy":"regulate","emotion":"outrage"},
{"id":"ytc_UgyaXW-dZhx1sWFwbtV4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"unclear","emotion":"disapproval"},
{"id":"ytc_Ugyp6IWnXQ8C4Cfvzxp4AaABAg","responsibility":"user","reasoning":"deontological","policy":"none","emotion":"mixed"},
{"id":"ytc_UgzRUHUAWGjGciAoLmh4AaABAg","responsibility":"user","reasoning":"deontological","policy":"none","emotion":"outrage"},
{"id":"ytc_UgzAfQzn9xFZAwWQu9B4AaABAg","responsibility":"user","reasoning":"deontological","policy":"none","emotion":"outrage"},
{"id":"ytc_UgzH_C6cj0c7lz9RwSh4AaABAg","responsibility":"user","reasoning":"deontological","policy":"none","emotion":"outrage"},
{"id":"ytc_UgxEmUVBCSX7Zxm_GGV4AaABAg","responsibility":"ai_itself","reasoning":"unclear","policy":"unclear","emotion":"fear"},
{"id":"ytc_UgxkN4Q32zWNT0Bx7wJ4AaABAg","responsibility":"unclear","reasoning":"unclear","policy":"unclear","emotion":"indifference"},
{"id":"ytc_UgyYrAq-42u-RZvREOJ4AaABAg","responsibility":"ai_itself","reasoning":"deontological","policy":"liability","emotion":"mixed"}
]

Raw LLM Responses