Raw LLM — Corpus Dashboard

Look up by comment ID

Random samples — click to inspect

G Humans were meant for more than just driving a truck. Let the A.I have these job… ytc_UgyFgvY9Q… G The industrial revolution may have given humans tools but for Horses it relaced … ytc_UgzwMJmT5… G Google: “We have no moat against open source AI” Open AI: “Grab a shovel and he… ytc_UgwyDVNv5… G Pretty scary conversation. My buddy says in five years there will be no big chan… ytc_UgyDLawfS… G 8:51. You said AI army when you meant to say military. You do a great job in the… ytc_UgxsPbLTK… G Terrifying. The only relief is that LLM is not AI in any way, despite propaganda… ytc_UgyW0VHgZ… G With the amount of censoring claude suffers, i cant imagine claude (anthropic) o… ytc_UgzzgXJI5… G At least AI companies now have a good deflection tactic. "It's not our product'… rdc_o7c3wvn

Comment

AI's don't have drives. He knows this.... he spends too much time reading predicted text and has deluded himself. It still is fancy autocomplete. Yes, there is other stuff bolted on top. Why is the AI saying how a patient reacted to epinephrine? That's nonsense without being able to observe the patient. Yes, it can guess, but so can I and so can a doctor. That's what autocomplete is, it's fundamentally detached from the physical reality of the situation. Solving math Olympiad problems is almost certainly training data contamination. It has been demonstrated that it takes very few examples to poison AI training for whatever output you desire in a narrow area. If AI were that great at math then it would be tremendously useful to me, and yet it isn't. It sucks hairy monkey balls at basically all math related problems I ask of it. Why? Because the problems I ask are novel and apparently the Olympiad problems were not this time around, they entered the training data somewhere to be regurgitated.

youtube AI Moral Status 2025-11-02T14:1… ♥ 1

Coding Result

Dimension	Value
Responsibility	ai_itself
Reasoning	consequentialist
Policy	none
Emotion	indifference
Coded at	2026-04-26T23:09:12.988011

Raw LLM Response

[{"id":"ytc_UgwoZ_ObFGWO8kS0MN94AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"indifference"},
{"id":"ytc_Ugxm8ymEkFJfTdvizG14AaABAg","responsibility":"developer","reasoning":"deontological","policy":"none","emotion":"mixed"},
{"id":"ytc_UgwXu4ZoKd5ie0rGLkp4AaABAg","responsibility":"developer","reasoning":"virtue","policy":"none","emotion":"mixed"},
{"id":"ytc_Ugxi90pefiwO-3ZJ75N4AaABAg","responsibility":"user","reasoning":"consequentialist","policy":"liability","emotion":"fear"},
{"id":"ytc_UgyMvq0VERxFUxZ9n5x4AaABAg","responsibility":"none","reasoning":"contractualist","policy":"none","emotion":"resignation"},
{"id":"ytc_UgzraAd2k9OgS67G7Ct4AaABAg","responsibility":"developer","reasoning":"deontological","policy":"regulate","emotion":"fear"},
{"id":"ytc_UgwhprLqk9khERGYPCx4AaABAg","responsibility":"company","reasoning":"virtue","policy":"ban","emotion":"outrage"},
{"id":"ytc_UgwccqDfXKUFdMg788V4AaABAg","responsibility":"none","reasoning":"mixed","policy":"none","emotion":"mixed"},
{"id":"ytc_UgxNNNjj3Wgf80ULMJ94AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"mixed"},
{"id":"ytc_UgzYqiRM8kumsn5QPgJ4AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"fear"}]

Raw LLM Responses