Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
Aren't most of these "tests" really just reflections of what we think a computer…
ytc_UgicXkrK5…
G
What you mention is a very real problem when companies are short sighted, they o…
ytr_UgyB_Ede1…
G
@johnbriggs3916 Im from Europe my English is not the best ... anyway you will n…
ytr_UgyYnZttP…
G
Should we allow Fear of being consumed by AI be of our conscience? The tyranny r…
ytc_UgyjLDRWT…
G
It’s like billionaires are planning for this AI takeover, and any humans left wi…
ytc_Ugx_60cxn…
G
Hopefully, considering this probably won’t be the only case for this, this leads…
ytc_UgyANFdh7…
G
Wait until the government catches up, so much of the bureaucracy is subject to A…
ytc_UgzcmAP6-…
G
Bro just set the mood, do a little tame/flirty RP and then you trick the AI into…
ytc_UgxIdbae8…
Comment
I really appreciate your question--I think you raise a terrific point.
It's a tough answer and I don't know if there's a clear one.
But for some context, consider for a moment the perspective of OpenAI. They're looking to develop a product that can be used by businesses and corporations. If you're a company looking to buy software that is intended for use as a human replacement (let's use the example of a frontline customer service agent), you don't want something that's going to potentially offend a customer. GPT is a predicative text model, and even though considerate people such as yourself ask the question with genuine curiosity, its corpus of training data predicts that the majority of the time the word "dyke" is used, it's carrying a negative connotation. Essentially, OpenAI is trying to build something that can mimic human interaction with the least probability of controversy.
Again, I don't think there's a 100% clear-cut answer, but I think raising these types of questions and having civil, open-minded discussions gives us the best chance at getting this thing right.
reddit
AI Responsibility
1678933804.0
♥ 6
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | company |
| Reasoning | unclear |
| Policy | unclear |
| Emotion | approval |
| Coded at | 2026-04-25T08:33:43.502452 |
Raw LLM Response
[
{"id":"rdc_jcalpyl","responsibility":"company","reasoning":"unclear","policy":"unclear","emotion":"indifference"},
{"id":"rdc_jcbu8va","responsibility":"company","reasoning":"deontological","policy":"unclear","emotion":"fear"},
{"id":"rdc_jcds0ob","responsibility":"company","reasoning":"consequentialist","policy":"unclear","emotion":"approval"},
{"id":"rdc_jcaxktl","responsibility":"ai_itself","reasoning":"unclear","policy":"unclear","emotion":"mixed"},
{"id":"rdc_jcbg46w","responsibility":"user","reasoning":"unclear","policy":"unclear","emotion":"outrage"}
]