Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
Don’t use ai even for fun Charlie. You seriously want to be seen as a piece of t…
ytc_UgymA9fD8…
G
Nobody is talking about the psychology of AI. Is it a psychopath or is it capabl…
ytc_UgzaYIhsp…
G
I personally find AI in general are both terrifying and fascinating at the same …
ytc_UgxaGrUNC…
G
La IA es la peor cosa que le pueda suceder al mundo.
Es una desgracia mayor que…
ytc_Ugx8BoROC…
G
I make diamond art, if youve never heard of it, its kinda like if mosiac and col…
ytc_Ugy-rQ6a1…
G
How dumb do you have to be to believe any of this garbage? Being anti AI is fine…
ytc_UgxSsaECA…
G
Honestly, he may be a nobel laureate, but he's dead wrong about LLMs. My only qu…
ytr_Ugww466wq…
G
Coders learn concepts at college and those concepts become intuition through wri…
ytc_UgyH9enfC…
Comment
Senator Sanders is the only politician I've heard talk about the dangers and problems with A.I. Even though Bernie is an American politician, This is an issue which is going to affect the entire world not just Americans. Recently for instance we only just discovered that the most advanced A.I's TODAY when ordered to shut down or destroy themselves they simply refuse to do it majority of the time and will do anything to avoid it. Some more than others but the fact it isnt 100% of the time alone is extremely concerning. One A.I actually blackmailed an employee involved in an experiment getting the A.I to shut down, Threatening to send an E-mail around a company to expose the fact they were having an affair.
Another test was done with another A.I to see how it would react to a human was trapped in a server room with the intention of shutting the A.I down. The A.I the vast majority of the time opted to let the person "expire" and not even contact authorities during or afterwards to say the person was in distress. This is with A.I's now, imagine how more advanced they are going to be in 5 or 10 years time, We may completely loose the ability to control them and right now big tech has absolute no answer to this major issue and is forcing this technology with breakneck speed with extreme recklessness and irresponsibility. Lessons from the past which harmed numerous people and almost lead to the end of civilisation during the cold war are being ignored or forgotten.
youtube
AI Jobs
2025-10-08T03:5…
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | distributed |
| Reasoning | consequentialist |
| Policy | regulate |
| Emotion | fear |
| Coded at | 2026-04-26T23:09:12.988011 |
Raw LLM Response
[
{"id":"ytc_Ugyi5ymzAgO0PU1BkDN4AaABAg","responsibility":"company","reasoning":"consequentialist","policy":"regulate","emotion":"approval"},
{"id":"ytc_UgwbEC6U0L4cTvCalBV4AaABAg","responsibility":"company","reasoning":"consequentialist","policy":"liability","emotion":"outrage"},
{"id":"ytc_Ugxo0lNp8HJ5CIA6oEF4AaABAg","responsibility":"government","reasoning":"deontological","policy":"none","emotion":"outrage"},
{"id":"ytc_Ugy8VmkeK3U0PEbjmnB4AaABAg","responsibility":"ai_itself","reasoning":"unclear","policy":"unclear","emotion":"fear"},
{"id":"ytc_UgybF1u3L7ShCxBoas54AaABAg","responsibility":"distributed","reasoning":"mixed","policy":"regulate","emotion":"approval"},
{"id":"ytc_UgydldKiOGEBBKa3scF4AaABAg","responsibility":"distributed","reasoning":"consequentialist","policy":"regulate","emotion":"fear"},
{"id":"ytc_UgyRJGw-EcRiRYacns14AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"unclear","emotion":"fear"},
{"id":"ytc_Ugwkfd2sskNfmPrx8at4AaABAg","responsibility":"company","reasoning":"consequentialist","policy":"regulate","emotion":"outrage"},
{"id":"ytc_UgyXXA5ZuvaugvrYwbN4AaABAg","responsibility":"company","reasoning":"unclear","policy":"unclear","emotion":"resignation"},
{"id":"ytc_UgxXv6DvOGJN-TmLX6t4AaABAg","responsibility":"distributed","reasoning":"consequentialist","policy":"regulate","emotion":"fear"}
]