Raw LLM — Corpus Dashboard

Look up by comment ID

Random samples — click to inspect

G so there's a category of Darwin award AI Darwin award also makes me not want to … ytc_Ugyhxt648… G Trying to say ai learns like a person is silly. It’s literally taking the data a… ytr_UgxmDl1hy… G As expected most of the arguments were placed on a political bias with the empha… ytc_UgwGRwroF… G We must end corporate control of America. We are slaves to a handful of people.… ytc_Ugz5JM6Yt… G @yourparalysisdemon1936 I don't think people are going to stop using AI art gene… ytr_UgxjP25VX… G Will AI replace IAS officers too?? I think maybe AI will make better policies fo… ytr_UgwthTFgr… G Fully automatic, deadly weapons, i.e. those who make decisions completely indepe… ytc_UgycByoLx… G The main problem with self driving cars isn't even technical or the fact they ar… ytr_UgznuBOtk…

Comment

Ezra, your question about an “off switch” for AI hits the core issue: control shouldn’t mean panic or shutdown, it should mean governance. The more realistic model isn’t a big red button — it’s a role and permissions system, like an operating system. Humans stay as root users; AIs run as processes with defined privileges. Instead of hoping we can unplug the machine, we architect clear boundaries: what the model can access, what data it can alter, and who approves escalations. That design treats AI as an augmentation layer, not a rival mind. It keeps humans as the final context — we decide scope, timing, and moral weight. In practice, the off switch isn’t a single command; it’s a security structure that scales with trust. When permissions are transparent and revocable, we get both safety and usability — the same principles that keep any complex system stable. We’ll provide more on this through time.

youtube AI Governance 2025-10-17T09:3…

Coding Result

Dimension	Value
Responsibility	distributed
Reasoning	contractualist
Policy	regulate
Emotion	approval
Coded at	2026-04-26T23:09:12.988011

Raw LLM Response

[
{"id":"ytc_UgzmexWnJbzB4UVydcp4AaABAg","responsibility":"developer","reasoning":"consequentialist","policy":"liability","emotion":"indifference"},
{"id":"ytc_Ugw7OLpNX_TZUxqq59p4AaABAg","responsibility":"unclear","reasoning":"mixed","policy":"unclear","emotion":"mixed"},
{"id":"ytc_UgwKEehWUnNlPWy_TWd4AaABAg","responsibility":"distributed","reasoning":"contractualist","policy":"regulate","emotion":"approval"},
{"id":"ytc_UgyU9nMB3UAMNASSNJJ4AaABAg","responsibility":"user","reasoning":"unclear","policy":"none","emotion":"indifference"},
{"id":"ytc_UgyP5g2sFlAM953W1SJ4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"resignation"},
{"id":"ytc_Ugx9H0IgRcbmLumw7BZ4AaABAg","responsibility":"unclear","reasoning":"deontological","policy":"none","emotion":"outrage"},
{"id":"ytc_UgwMPclbDSD7WueoaUd4AaABAg","responsibility":"developer","reasoning":"deontological","policy":"regulate","emotion":"fear"},
{"id":"ytc_UgyS6eg3Ahxh9j_h0xl4AaABAg","responsibility":"unclear","reasoning":"unclear","policy":"unclear","emotion":"outrage"},
{"id":"ytc_Ugza-ErPaJCR14qaidV4AaABAg","responsibility":"developer","reasoning":"consequentialist","policy":"unclear","emotion":"indifference"},
{"id":"ytc_UgyqpjoKAD_xVT18qRh4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"fear"}
]

Raw LLM Responses