Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
The default trajectory is highly dangerous. We the general public can do somethi…
ytr_UgwRzp1RM…
G
Where is Will Smith when ya need him‼️ (Guess we didn’t learn from i robot)…
ytc_UgyW3VpYe…
G
Explanation: When he said "not human kids", Meta AI drew a picture of two black …
ytc_UgwKKEjGp…
G
Okay. So we pivot from unaccountable tech billionaires to now their equally una…
ytc_UgxVNqlku…
G
I wish you could fuck with the training programs by hiding code in the image and…
ytc_UgwL0Sk7F…
G
@31:00 he talks about control.
The only thing anyone has control over in this …
ytc_Ugw-ZL4FE…
G
ChatGPT is programmed to spit out consensus views on many subjects.If you point …
ytc_UgxUlQUZZ…
G
like even the real estate company my dad works in is using ai. like why is ts ha…
ytc_Ugxmz21LA…
Comment
I think what most people are trying to use AI for is fundamentally wrong. They aren't fact machines. They're more like impressionable children. Sponges that soak up ideas. That's just in the very nature of the neural network approach.
Just like a human, you can't make them perfectly impartial. You first have to teach them right from wrong, and then train them to try to compensate for their personal biases.
It sounds kooky, but AI IS kooky.
reddit
AI Harm Incident
1743248826.0
♥ 1
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | distributed |
| Reasoning | virtue |
| Policy | none |
| Emotion | resignation |
| Coded at | 2026-04-25T08:33:43.502452 |
Raw LLM Response
[
{"id":"rdc_mk81dza","responsibility":"distributed","reasoning":"consequentialist","policy":"liability","emotion":"fear"},
{"id":"rdc_mk8i9h9","responsibility":"company","reasoning":"deontological","policy":"liability","emotion":"outrage"},
{"id":"rdc_mk91kr0","responsibility":"user","reasoning":"mixed","policy":"industry_self","emotion":"approval"},
{"id":"rdc_mkbceh6","responsibility":"company","reasoning":"consequentialist","policy":"regulate","emotion":"indifference"},
{"id":"rdc_mkcbfcm","responsibility":"distributed","reasoning":"virtue","policy":"none","emotion":"resignation"}
]