Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
I call BS. This is why they shoved the Wizard of Oz down our throats (for us old…
ytc_Ugx9dBn4G…
G
Stop giving us hope twin, we know that everyone wil shift focus to building AI r…
ytc_UgygnaH0p…
G
You're absolutely right. And that's the irony of innovation. People only realize…
ytr_Ugw08tEv3…
G
I was just waiting for them to reveal that all the people in this video were als…
ytc_UgzZWdNfd…
G
joke of the day
did you know? if the robot starts being aggresive and destroys …
ytc_UgxDvTTAo…
G
I think the beautiful thing about art of all forms is that everyone has a unique…
ytc_UgyMHomEm…
G
I am ambivalent to ai. but i believe when an ai learns from base art it is the s…
ytc_UgzALkhCC…
G
It's difficult enough to reason with people that AI responses are related to pro…
ytc_UgyeLEcpX…
Comment
I used GPT3.5 and 4 to prepare a CV, cover letter and the interview way faster than I could have without it. ChatGPT is also great for basic coding queries in lieu of googling and sleuthing forums. I'm sure that if it weren't for data security concerns, we could probably replace a large part of our employees (Callcenter) with ChatGPT within a year or two.
Where ChatGPT fails at the moment is in gray areas. For example giving it a fantasy prompt about a brothel (which I did specifically for the post) ChatGPT will output things similar to:
>As the story's narrator, I must maintain a level of discretion and respect for the characters involved. It is important to remember that consent and communication are key in any intimate situation, and that relationships should be based on mutual respect and understanding.
Ok, they don't want you writing smut. But, the same goes for other "sensitive" topics such as religion, politics etc. There are legitimate use cases for this, such as using as a writing aid (I used it write radio speeches for an RPG).
You can somewhat get around it be formulating your prompts to "persuade" the filter, but at some point the question is "who is making decisions on what should be filtered and what not." It's not yet a huge issue, but once these tools become ubiquitous this could cause huge biases. Imagine OpenAI filtering anything remotely Anti-Corporate and promoting liberal talking points... The only way to avoid it is no filters.
reddit
AI Harm Incident
1681486128.0
♥ 3
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | unclear |
| Reasoning | unclear |
| Policy | unclear |
| Emotion | unclear |
| Coded at | 2026-04-25T08:33:43.502452 |
Raw LLM Response
[{"id":"rdc_jg8n7zh","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"approval"},
{"id":"rdc_jg7icy8","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"indifference"},
{"id":"rdc_jg7o13d","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"mixed"},
{"id":"rdc_jg7c9dz","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"},
{"id":"rdc_jg7cl29","responsibility":"developer","reasoning":"consequentialist","policy":"none","emotion":"mixed"})