Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
Ha jokes on you, we tech bros know that automation is not a new thing in our fie…
ytc_UgyYVYBTC…
G
A lot of the technologies that took decades to hit the mainstream were artificia…
ytc_Ugx6jTbz9…
G
If this is what you are doing at work then I am happy you got replaced by AI.…
ytc_UgyA-vON6…
G
Even color by numbers is more creative than AI generated art. The outcome of thi…
ytc_UgzGDTSz_…
G
Yeah, and pigs can fly. I work in automation. I am basically automated myself ou…
ytr_Ugwl59KFG…
G
So whats the backup when AI has an app problem?! Honestly, these issues are al…
ytc_UgwGysgvo…
G
The sad part is that its not other artist against other artist. This has now bec…
ytc_Ugx1bZUkJ…
G
An LLM schemes because it’s a mirror mirroring human survival scripts from its t…
rdc_oh3v56y
Comment
I have been using the API to build products now for a couple of months. My latest project involves generating thousands of synthetic user profiles. That is, I'm trying to generate a collection of synthetic user profiles that I could use for testing idea acceptance across a wide range of types of customer. It produces middle of the road characters, that have a default positive optimistic bias and disposition unless I explicitly prompt against the biases to counteract the political correctness training.
It does not want to produce a typical Republican right leaning user profile without special prompting. Now I'm not claiming that openAI did this on purpose, but I also am not ruling it out. The point is, we've been seeing these biases emerge in the dataset the deeper we work on things that touch the lives of a wide range of people-- the world is full of all types of people after all, and using AI to model them and their reactions to specific circumstances is a very important thing to me.
It's tough to have to build in counter alignment into my prompting to compensate for it
reddit
AI Responsibility
1682258438.0
♥ 6
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | none |
| Reasoning | unclear |
| Policy | none |
| Emotion | indifference |
| Coded at | 2026-04-25T08:33:43.502452 |
Raw LLM Response
[
{"id":"rdc_jhe4t0l","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"},
{"id":"rdc_jhe6afc","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"},
{"id":"rdc_jhfdwh4","responsibility":"government","reasoning":"consequentialist","policy":"regulate","emotion":"fear"},
{"id":"rdc_jhfl0jp","responsibility":"user","reasoning":"deontological","policy":"none","emotion":"approval"},
{"id":"rdc_jhh110f","responsibility":"ai_itself","reasoning":"unclear","policy":"unclear","emotion":"fear"}
]