Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
집안 탁자에 올려 놓으세요. 그러면 도둑은 못 들어 올 겁니다. "Just leave it on the table at home. That wa…
ytc_Ugyqyrjuo…
G
The danger of hyper intelligent AI? Us… we are hyper intelligence and look at us…
ytc_UgyEaI4bP…
G
Anyone can use a camera, yes, but it takes technique to take a picture like a ph…
ytr_UgwLbSJdO…
G
Of course we know what goes in inside these models. Stop the fearmongering alrea…
ytc_UgyrO0-H4…
G
Am i the only one with doesn't hate ai to much even though im a artist? -…
ytc_UgxECl-tN…
G
She is real, I asked how much and just like a real woman not enough.…
ytc_UgzAwrkEg…
G
Man asks AI for health advice, his health gets significantly worse, shock of the…
ytc_UgwrMzrM1…
G
Unfortunately, the original prompt forced ChatGPT to avoid your implied actual g…
ytc_UgxP8qwpc…
Comment
The easiest way is Google Docs. As a high school teacher, I use an extension that shows the letter by letter typing history. It works well because human typing of a long assignment is very different from even typing a copy of an AI answer. You naturally tend to go back and forth, delete chunks of text and rewrite, misspell and fix, etc. If a high schooler writes a perfect paragraph in one go with none of that, with perfect spelling and grammar, it's very sus. It helps that I know what their base writing capabilities are.
As such, the best option for you also is to write answers in google docs and then send them or copy/paste into the answer box.
reddit
AI Harm Incident
1765778181.0
♥ 68
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | user |
| Reasoning | consequentialist |
| Policy | industry_self |
| Emotion | approval |
| Coded at | 2026-04-25T08:33:43.502452 |
Raw LLM Response
[
{"id":"rdc_nu1uvam","responsibility":"none","reasoning":"unclear","policy":"unclear","emotion":"indifference"},
{"id":"rdc_nu290r5","responsibility":"none","reasoning":"mixed","policy":"none","emotion":"mixed"},
{"id":"rdc_nu25jd5","responsibility":"user","reasoning":"consequentialist","policy":"liability","emotion":"fear"},
{"id":"rdc_nu42ch6","responsibility":"user","reasoning":"consequentialist","policy":"industry_self","emotion":"approval"},
{"id":"rdc_nu397fl","responsibility":"company","reasoning":"consequentialist","policy":"regulate","emotion":"outrage"}
]