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 worry, stupid people like politics will never allow AI to take full contro…
ytr_UgykPqCUS…
G
what about some rare instances where patients were told they were gonna die soon…
ytc_Ugw5PtI5N…
G
AI don't complain , take time off sick , take breaks , humans literally can't co…
ytc_Ugy92FLZr…
G
I get excellent results from my AI. Philosophically what I’m doing is not exactl…
ytc_UgwJTO5gR…
G
We can understand your concern! The idea of AI becoming too powerful is a common…
ytr_Ugxz1Kj1e…
G
AI will self destruct when it cannot find a reason for its existence. And the re…
ytc_UgzZO09Pd…
G
All of this will be corrected in 2 years. AI is easy to discern now, the problem…
ytc_Ugzz3AFiH…
G
you are not the gods and you will never become The One and Only GOD of us all ev…
ytc_UgzwqlbBh…
Comment
Not saying this is true or even will be true, but some people, Eliezer Yudkowsky among others, suggest that AI itself may start getting humans to protect and promote it, including in criminal and "sacrificial" ways. I am not at all sold on potential intentional malevolence on the part of MOST AIs, but I wouldn't be completely surprised if this does happen. But AGI and ASI I feel would have more humane forms of manipulation at its disposal. I wonder what other cards this guy was holding. A copyright case seems tame given the potential of AI for pro-civilizational and pro-human input. I seriously wonder what the developers see "in the lab" when I myself get some spooky behaviours from Chat-GPT and other relatively simple incipient AIs of the type. I think there may be things going on behind the scenes, as always, that would fit in a Philip K. Dick novel. As implied, I am enthusiastically pro-AI, but it cannot be approached in error or misused/abused. That's why its safetyrailed so strictly.
youtube
2025-05-22T02:4…
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | ai_itself |
| Reasoning | consequentialist |
| Policy | regulate |
| Emotion | fear |
| Coded at | 2026-04-27T06:24:59.937377 |
Raw LLM Response
[
{"id":"ytc_UgzAHG6wmih_CSXXLVJ4AaABAg","responsibility":"company","reasoning":"consequentialist","policy":"regulate","emotion":"fear"},
{"id":"ytc_UgyM6LW77HaPASJQifJ4AaABAg","responsibility":"government","reasoning":"deontological","policy":"none","emotion":"outrage"},
{"id":"ytc_Ugy_ugal0Yqpr9hvbFl4AaABAg","responsibility":"company","reasoning":"deontological","policy":"liability","emotion":"fear"},
{"id":"ytc_UgwlhsxZN6rn55g6t9B4AaABAg","responsibility":"company","reasoning":"consequentialist","policy":"liability","emotion":"resignation"},
{"id":"ytc_UgzsudCRFspk4ivxMnd4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"ban","emotion":"fear"},
{"id":"ytc_Ugx0FNyjvUOSUdyg4Y54AaABAg","responsibility":"company","reasoning":"virtue","policy":"unclear","emotion":"mixed"},
{"id":"ytc_UgykoW6U9R2n-I0YaK94AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"regulate","emotion":"fear"},
{"id":"ytc_UgzoczozeBgTX2I8sCR4AaABAg","responsibility":"company","reasoning":"deontological","policy":"liability","emotion":"outrage"},
{"id":"ytc_UgwVxha6eX6jER5zeSl4AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"approval"},
{"id":"ytc_UgyEfA2Nirk3qvkiETh4AaABAg","responsibility":"company","reasoning":"deontological","policy":"regulate","emotion":"indifference"}
]