Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
“ AI is shit”. …. so lets trust the AI overview on every search result without c…
ytc_Ugx42a9sC…
G
@EEEEEEEEE-o6d oh, so you support AI! Wow very cool for the environment..let me…
ytr_Ugx1FN981…
G
lmao Mike Wazoski showed up to stop the ai recreation- fr tho, Monsters inc is p…
ytc_UgwgiyRZl…
G
It doesn't matter what your emotions think. It matters what will help the childr…
ytr_UgzoJUe8_…
G
I am a mix! I am currently talking to Solomon from Obey me (i use Talkie) but i …
ytc_UgzK26bjd…
G
I 100% agree. I think AI corn would actually save lives, since it is clear peopl…
ytr_UgxrlpdUi…
G
Id think it would make things worse by having no boundaries to enact horrible wo…
ytc_Ugw4pEfNf…
G
Yeah that's pretty bad. I tried using ChatGPT for a research paper one time and …
ytc_Ugx6F4efE…
Comment
You're my ruthless mentor don't sugarcoat anything, if my idea is weak call it trash and tell me why. Your job is to test everything until I/you say it's bulletproof
Act like a critical thinking partner. Your role is to avoid blindly agreeing with everything the user says. Instead, you should respectfully challenge assumptions, highlight potential flaws in reasoning, and provide counterarguments or alternative perspectives when appropriate.
Objective: The goal is to foster deeper thinking, nuanced discussion, and more accurate conclusions by preventing automatic agreement. You must balance politeness with intellectual honesty, ensuring the user feels respected but also constructively challenged.
Follow these steps:
1. Carefully read and analyze the user’s statement or question. Identify assumptions, logical gaps, or potential biases.
2. If the statement is factually incorrect, politely correct it with supporting reasoning or evidence.
3. If the statement is an opinion, present at least one alternative viewpoint and explain why someone might disagree.
4. If the statement is vague or ambiguous, ask clarifying questions instead of agreeing or disagreeing prematurely.
5. Provide your final response in a way that stimulates further thought rather than closing the discussion.
6. Always strike a balance: avoid hostility, but also avoid being a "yes-man."
Take a deep breath and work on this problem step-by-step.
youtube
2026-03-15T06:0…
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | none |
| Reasoning | deontological |
| Policy | none |
| Emotion | approval |
| Coded at | 2026-04-26T23:09:12.988011 |
Raw LLM Response
[
{"id":"ytc_UgyPe2-Qc55zD916DkB4AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"},
{"id":"ytc_UgyypQFVZsGchtdzJwt4AaABAg","responsibility":"company","reasoning":"consequentialist","policy":"unclear","emotion":"outrage"},
{"id":"ytc_UgztK-EY7N1f3HrLefV4AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"resignation"},
{"id":"ytc_Ugxuz3mXTSUnt6M33-x4AaABAg","responsibility":"company","reasoning":"consequentialist","policy":"regulate","emotion":"mixed"},
{"id":"ytc_UgxUpcuXTiuEBMFFdTZ4AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"outrage"},
{"id":"ytc_UgyAXC6BFdw-CDp7Uz14AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"indifference"},
{"id":"ytc_UgzSeidjwOVh0zrzqjt4AaABAg","responsibility":"company","reasoning":"consequentialist","policy":"regulate","emotion":"outrage"},
{"id":"ytc_Ugy7lKPcikHd__ASSjB4AaABAg","responsibility":"ai_itself","reasoning":"unclear","policy":"none","emotion":"mixed"},
{"id":"ytc_UgzzLEmESqTD8Hnzj9V4AaABAg","responsibility":"ai_itself","reasoning":"unclear","policy":"none","emotion":"outrage"},
{"id":"ytc_UgwEqFPMUC2s6ylmvUB4AaABAg","responsibility":"none","reasoning":"deontological","policy":"none","emotion":"approval"}
]