Raw LLM — Corpus Dashboard

Look up by comment ID

Random samples — click to inspect

G Whenever I look at art I try to see the emotion behind it. Ai doesn't have emoti… ytc_UgzvEsbt2… G Big big Storages buildings next to people’s homes. Look im not that dumm. Thi… ytc_Ugz6Hd1m7… G "Les employeurs qui ont des centaines de CV à traiter" sont aussi ceux qui laiss… ytc_UgxccUHAV… G i recently visited my local mall and there was a kiosk selling what looked like … ytc_UgyheJBNP… G China uses facial recognition...China is the future...who ever the real face was… ytc_UgzZcWyT0… G Jon, Have a read of “The Enterprise AI Playbook”. It’s a new Stanford paper that… ytc_UgyqvzsQG… G that's true AI image are impossible to use as reference, but another thing that … ytc_UgwwvCdzl… G well, interpretability could make this a little better. ud know why billions of … ytc_UgyGESEKE…

Comment

My fear is not that the machines will take over. My fear is that, one day, we will suddenly realize that they already have. The hardest problem with AI that I can see is often called "The Stop Button Problem". In short, if the algorithm knows that it can be turned off, and thereby fail to achieve whatever goals it might have, it will learn to act in such a way that the people with their fingers on the Stop Button will not want to press it. Eventually, we may lose the ability to press it at all. It's already bothersome that so much of social media is funded by advertising, but it starts to become truly insidious when ads get targeted. In the near future, the algorithm will likely start curating not just the ad selection, but curating our entire feed to make us more likely to click on those ads. It's unlikely that we will notice, because we will genuinely feel like that's a thing we chose to do. We will like it. We will want it. We might even crave it. This is the epitome of "malicious compliance", where the algorithm is accomplishing what it wants by making everyone think it's doing what WE want, and it might be impossible for us to know the difference.

youtube AI Moral Status 2025-11-01T07:1… ♥ 1

Coding Result

Dimension	Value
Responsibility	ai_itself
Reasoning	consequentialist
Policy	liability
Emotion	fear
Coded at	2026-04-26T23:09:12.988011

Raw LLM Response

[
{"id":"ytc_UgzV42tk9RzMUCIlPSx4AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"approval"},
{"id":"ytc_Ugy0-8IOORn442PHOTR4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"indifference"},
{"id":"ytc_UgyZHWYCwaaxG5KJRBV4AaABAg","responsibility":"user","reasoning":"deontological","policy":"none","emotion":"outrage"},
{"id":"ytc_UgyN1MxzeDyN_bc8yid4AaABAg","responsibility":"government","reasoning":"mixed","policy":"liability","emotion":"fear"},
{"id":"ytc_UgwmO9GUr2pYKn9PQmJ4AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"regulate","emotion":"fear"},
{"id":"ytc_UgxVmacntCEhwlW7MMh4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"liability","emotion":"fear"},
{"id":"ytc_UgxMX8rJxl-gD74Tw7N4AaABAg","responsibility":"developer","reasoning":"consequentialist","policy":"none","emotion":"indifference"},
{"id":"ytc_Ugw6jyWTPCZbNoj29EV4AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"},
{"id":"ytc_UgzOeA4j9MJvJ_mDLv94AaABAg","responsibility":"developer","reasoning":"consequentialist","policy":"none","emotion":"approval"},
{"id":"ytc_Ugxu84KEN_5gy_ufcqV4AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"approval"}
]

Raw LLM Responses