Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
Question for Casey: If you tell AI to delete system32 and it works are you liabl…
ytc_UgxCUAMcQ…
G
You most likely gonna ignore this and am no layer but I suggest you to collect e…
ytc_UgyMhtnhk…
G
I could hear a guy laughing I mean come on it’s not a joke. They don’t understan…
ytc_UgwCJ7KnH…
G
Hard to clearly define the down sides ??? how about AI drone bombing and and gen…
ytc_UgzpgEcHW…
G
Society benefits more when it runs efficiently. We could extend prison sentences…
rdc_czlwiqs
G
How about having the robots do the job and let the humans get the paycheck. I me…
ytc_UgyJRaz7y…
G
to everyone being negative, realize that sam isnt a God. Deepseek, by itself, ru…
ytc_Ugy8fk-Po…
G
No you can't make AI safe once it reachesv super agi, didn't you watch Terminat…
ytc_UgxUphiGN…
Comment
100% of the code generated by GPT-4 in their analysis is executable if they actually bothered to parse it.
[Source](https://www.reddit.com/r/ChatGPT/comments/153xee8/has_chatgpt_gotten_dumber_a_response_to_the/)
reddit
AI Harm Incident
1689779403.0
♥ 11
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | none |
| Reasoning | consequentialist |
| Policy | none |
| Emotion | indifference |
| Coded at | 2026-04-25T08:33:43.502452 |
Raw LLM Response
[{"id":"rdc_jskfpeo","responsibility":"ai_itself","reasoning":"mixed","policy":"none","emotion":"outrage"},{"id":"rdc_jslk9i2","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"indifference"},{"id":"rdc_jslntdr","responsibility":"company","reasoning":"deontological","policy":"none","emotion":"outrage"},{"id":"rdc_jskmscy","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"mixed"},{"id":"rdc_jsldrtq","responsibility":"company","reasoning":"consequentialist","policy":"none","emotion":"outrage"}]