Raw LLM — Corpus Dashboard

Look up by comment ID

Random samples — click to inspect

G Anything with CHINA IS EVIL and S A T A N IC Based. That Fiery Red Dragon Satan… ytc_UgwrJtXjK… G Eh, I like the kind of AI pictures that are obviously AI-made. People just don't… ytr_UgxJy1RYq… G Knew this was going to happen. Even YouTube has AI generated Medicare ads that … ytc_Ugy4x2cVP… G You have it half right, the culling is conscription into world war 3 to flatten … ytr_UgzLEK2hA… G well, cruise missiles already run an AI pilot. this is just one step up from th… ytc_Ugx2InJIz… G No but the creating experiments and interpretating results can be. If facial rec… ytr_UgwfqtV2-… G If President Trump stop the tariffs and start making better trades with allies f… ytc_UgxrvYJ3a… G Same. And I can’t use hyphens or anything like that anymore since apparently AI … ytr_UgzIdEpWD…

Comment

@redmint4894 I guess I used the word intuitive too often, will correct it in the text. I think it's more about patterns in the data, when there's f.e. more and stronger direct associations of teenage girls with Beyonce and make up in the data, the LLM gets stronger connection strengths there. On that basis it intuits from specific prompts that he must be a teenage girl. This can be corrected f.e. by telling the LLM that it got it wrong, and other methods to tackle bias. When in our culture cats get attributed properties that match more learned representations of what is female than what is male, we intuitively come to conclusions. It's not a logical process and such biases in people are very difficult to confront and change. Not impossible but more difficult than making corrections in deep learning machines. I keep posting a link to a video of a recent public lecture where he talks about discrimination and bias (from 46.12 to 50 minutes in the video), but the youtube algorithm constantly seems to remove it https://youtu.be/rGgGOccMEiY?t=2772

youtube AI Governance 2023-07-02T18:4… ♥ 1

Coding Result

Dimension	Value
Responsibility	developer
Reasoning	consequentialist
Policy	regulate
Emotion	approval
Coded at	2026-04-27T06:26:44.938723

Raw LLM Response

[
{"id":"ytr_Ugx9cKosGmQfk4IwBHp4AaABAg.9tnUaNc7BNQ9tr9MHH5_zv","responsibility":"company","reasoning":"consequentialist","policy":"regulate","emotion":"approval"},
{"id":"ytr_UgwgK26FkEvG2Yi4z6x4AaABAg.9tbxSjbHTrv9tcbHiXk0n4","responsibility":"ai_itself","reasoning":"consequentialist","policy":"liability","emotion":"fear"},
{"id":"ytr_UgwgK26FkEvG2Yi4z6x4AaABAg.9tbxSjbHTrvA4J_uwZpVua","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"},
{"id":"ytr_UgwgK26FkEvG2Yi4z6x4AaABAg.9tbxSjbHTrvA6_diDg_6VF","responsibility":"user","reasoning":"deontological","policy":"industry_self","emotion":"approval"},
{"id":"ytr_UgwgK26FkEvG2Yi4z6x4AaABAg.9tbxSjbHTrvA7UoS_qdUEg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"mixed"},
{"id":"ytr_UgwgK26FkEvG2Yi4z6x4AaABAg.9tbxSjbHTrvA7Ut55eGfeH","responsibility":"developer","reasoning":"consequentialist","policy":"liability","emotion":"fear"},
{"id":"ytr_Ugyf4MFBQogkCMSMKkJ4AaABAg.9sue6eMeOhT9t1MaN6U1-i","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"approval"},
{"id":"ytr_UgzCJ8fXQ8-Dz2NfNop4AaABAg.9rdPpOgAzW19rf4VgBMfm1","responsibility":"none","reasoning":"virtue","policy":"none","emotion":"indifference"},
{"id":"ytr_UgzCJ8fXQ8-Dz2NfNop4AaABAg.9rdPpOgAzW19rf7Zcocttd","responsibility":"none","reasoning":"deontological","policy":"none","emotion":"outrage"},
{"id":"ytr_UgzCJ8fXQ8-Dz2NfNop4AaABAg.9rdPpOgAzW19rfeHPWpZfs","responsibility":"developer","reasoning":"consequentialist","policy":"regulate","emotion":"approval"}
]

Raw LLM Responses