Raw LLM — Corpus Dashboard

Look up by comment ID

Random samples — click to inspect

G an AI wouldn't be woke since homophobia, transphobia, racism etc. are all human … ytr_UgzSso9vp… G Has anyone asked A.I. if it, well, 'thinks', it's intelligence is artificial? Wh… ytc_UgzDoqwPA… G I don't see what the drama is about. JUST GET OUT OF THE WAY, and A.I., has no r… ytc_UgybagFKc… G My question is: Why are we not publicly naming these powerful people that are ly… ytc_UgyWigPDb… G So what happens when you're going down the road at 70 mile an hour?And I don't k… ytc_Ugx4e4i2v… G I used to be an ai 'artist'. I didn't say it was mine or something but I would u… ytc_UgzcoVuYz… G I'm sure when people learn the real context of the story, it will lose all of th… ytc_UgwOFswGL… G Don't worry, the future is in good hands! Our AI models are here to assist and m… ytr_Ugx4IMW92…

Comment

@horizon-th2392 Believe me, I've watched his entire four part series many times, and I know about how the weights are continuously updated with gradient descent. It's a beautiful piece of math, no doubt, and as Mathematics enthusiast it is a miracle this technology was discovered. But the interpretability of such models is incredibly difficult for anyone. Additionally, I think a major flaw in machine learning is the absence of a strong theory of distributions as a foundation. This makes them more flexible, but it also makes it so you can't run vigorous tests of significance. Coming from a statistical background, it is standard that when you design a model that tests for the significance of parameters in your model. If you add parameters that are insignificant into your model, these tests will tell that you should remove them. This helps prevent overparameterizing a model. No such formal tests exist in a neural network. More simply, people rely on minimizing a criterion, like mean squared error. These are useful metrics, but so much of the rigor that exists in traditional statistical models is absent from these neural networks. When you cannot test if a neuron's presence is significant within a neural network, then you have to guess and check. Same with adding additional hidden layers. This is so unscientific it hurts. It's shame and neural networks need so much more research behind them before we can consider using them for tasks that have such serious implications.

youtube AI Harm Incident 2021-05-11T17:4… ♥ 1

Coding Result

Dimension	Value
Responsibility	none
Reasoning	unclear
Policy	none
Emotion	mixed
Coded at	2026-04-27T06:26:44.938723

Raw LLM Response

[
{"id":"ytr_UgyHcC6ZbGWv0kxXfPJ4AaABAg.AKyspzFigsUAKzeWSzLItd","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"fear"},
{"id":"ytr_UgyHcC6ZbGWv0kxXfPJ4AaABAg.AKyspzFigsUAL72oKDMCQV","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"},
{"id":"ytr_Ugyuba4POsJGww0kbct4AaABAg.9AnTx5CfyWM9SoUBTkUZR2","responsibility":"company","reasoning":"deontological","policy":"liability","emotion":"outrage"},
{"id":"ytr_UgyDzigDi5vQJ292Jg94AaABAg.9A5muLwgrKP9N9tb4Gxm6i","responsibility":"company","reasoning":"consequentialist","policy":"regulate","emotion":"fear"},
{"id":"ytr_UgyDzigDi5vQJ292Jg94AaABAg.9A5muLwgrKP9NBXaRVUk9E","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"approval"},
{"id":"ytr_UgyDzigDi5vQJ292Jg94AaABAg.9A5muLwgrKP9NCxfit8KmE","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"mixed"},
{"id":"ytr_UgytKvBE8iu4WADcwIh4AaABAg.99x5AbHg1Am9K61ilhqMGU","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"},
{"id":"ytr_UgzckBguyTLTh88eaXF4AaABAg.98lWrymEBwO99zTucip-1O","responsibility":"company","reasoning":"deontological","policy":"liability","emotion":"outrage"},
{"id":"ytr_Ugz-8ldChY_OOacRLGV4AaABAg.98FTJAG_P4v99wLNo4tnbj","responsibility":"company","reasoning":"deontological","policy":"regulate","emotion":"outrage"},
{"id":"ytr_UgxfrqjJS_wlu8ONq1J4AaABAg.97wuUI8xqkt99xH-WHX6l3","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"mixed"}
]

Raw LLM Responses