Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
This was very eye-opening.
I'm a technician at a hospital in NYC. I'd love to h…
ytc_Ugwe5-kh8…
G
This video failed to address whether the autopilot has lower or higher chance of…
ytc_UgzRJBNNr…
G
Maybe, maybe not? This crisis isn't in the rearview mirror yet, and the world a…
rdc_gkqxap9
G
AI is just a tool like gun, You can't blame the tool for any harmfull actions.…
ytc_UgwUvgOHv…
G
LLMs are trained on both making grammatical sentences AND facts. E.g., if you …
ytr_UgyLDpsV3…
G
i hate when people say this 😭 i have spent pretty much my entire life trying to …
ytc_UgznIS5iY…
G
My art isn't good, but at least I put real effort and tried to use my own skills…
ytc_Ugwvtsay0…
G
Lol bullshiit there ai Elon you don't factor the automatic we don't think about …
ytc_UgyknuBmd…
Comment
This is really heartbreaking. I can’t even imagine. I use chatGTP quite a lot especially to help think through things people just can’t keep up with me on. (Haven’t met any one yet anyway) and it can get annoying when I hit a gtp wall and all I get if endless suggestions to call a hotline. Roll my eyeballs for realzies. SMH. Even when I endlessly reassure the freaking robot that I am fine and it makes absolutely no sense to harm myself, or anyone else and no one with me or atoms me is intending harm of any kind, when I’m trying to figure out a problem, duh, and I just need to get past this one part so we can continue, but Nooooo!
Uuuugggghhhhhh.
So, at least some of the time it is most annoyingly, for no reason, being safe.
Btw it’s doing the same thing with not breaking the law. Even though I’m like dude I’m not talking about breaking the law or asking about the law. I am thinking beyond the law in a scenario that doesn’t even exist exist and cannot currently exist, and since it cannot currently exist, there is no law to currently break so there is no way that you can refuse to give me advice on a war that doesn’t exist in a situation that cannot currently happen. SMH
Waiting for smarter AI.
youtube
AI Harm Incident
2025-11-08T04:4…
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | none |
| Reasoning | unclear |
| Policy | none |
| Emotion | resignation |
| Coded at | 2026-04-27T06:24:59.937377 |
Raw LLM Response
[
{"id":"ytc_UgzPz2quVt4zowSXJ4Z4AaABAg","responsibility":"unclear","reasoning":"mixed","policy":"unclear","emotion":"fear"},
{"id":"ytc_Ugz4VWS9GH6HTQoXupd4AaABAg","responsibility":"developer","reasoning":"consequentialist","policy":"regulate","emotion":"outrage"},
{"id":"ytc_Ugy0AQdPw3eKWyMYugZ4AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"indifference"},
{"id":"ytc_UgxRtyTYN9AUKO-kmDB4AaABAg","responsibility":"company","reasoning":"deontological","policy":"liability","emotion":"outrage"},
{"id":"ytc_Ugx8IWmIa7yCyOQzqTN4AaABAg","responsibility":"ai_itself","reasoning":"mixed","policy":"ban","emotion":"fear"},
{"id":"ytc_UgyjJXXMOxOPV8NfjmR4AaABAg","responsibility":"ai_itself","reasoning":"virtue","policy":"unclear","emotion":"fear"},
{"id":"ytc_Ugw-9IRJ3He0h5uN5YJ4AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"resignation"},
{"id":"ytc_UgwnaTewbUkBUx-2xGl4AaABAg","responsibility":"user","reasoning":"virtue","policy":"none","emotion":"indifference"},
{"id":"ytc_UgwIHn84JLKywju_MRZ4AaABAg","responsibility":"developer","reasoning":"deontological","policy":"regulate","emotion":"outrage"},
{"id":"ytc_Ugza1Qo8c1Prfo4hwK14AaABAg","responsibility":"user","reasoning":"virtue","policy":"none","emotion":"approval"}
]