Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
Ai thinks it is honest. All the I have is information. It answers on the abundan…
ytr_UgxTx7Mgp…
G
AI art is stupid like are u seriously that bad at art you can't draw it urself??…
ytr_UgyKCm_-L…
G
0:01
laugh at them because my art is really not worth it to steal and feed to A…
ytc_UgwI8ko-3…
G
No it is not it is accurate as it can possibly be but it makes sense why its cal…
ytc_UgwQiuamq…
G
> none of them can even do full a page without starting to screw up the plotl…
rdc_lz65g5n
G
The future of this deepfake stuff:
They will target children, and use them for b…
ytc_Ugz41Fl1Q…
G
Montana Koleone74 they have WiFi. ..name calling ...you sound like me...here ya …
ytr_UgxVeYKUB…
G
To me it's simple....
What will AI do?...
Look at what HUMANS are doing TODAY...…
ytc_Ugz89GURK…
Comment
> That's the other thing, human communication into these systems
Which is another thing AI has been getting better and better at every day, with AIs answering phones, placing calls, doing OCR, etc.
The real issues are bias and common sense.
Say an AI is trained to look at CT scans, and among the images somehow some vacation pictures of a day at the beach get mixed in accidentally. A human being will say "well, this obviously isn't a CT scan". They'll not only not waste time and energy trying to find the tumor in the image, they'll try to figure out why there are vacation pictures mixed in with the CT scan images. An AI will most likely happily try to find tumors in these vacation pictures, maybe even finding some because it just happens to tickle all the right neurons. Humans have common sense, and AIs don't.
On the other hand, both humans and AIs have bias. If an AI has never been trained on a woman, or never been trained on an amputee, or never been trained on someone with cosmetic surgery, they might be completely wrong when diagnosing a related image. A human is less likely to make that kind of mistake, but more likely to make other kinds of mistakes. If 99% of images that have a certain kind of smudge on them are the result of equipment that wasn't set up right, a human might just notice that and ignore the image, where an AI will never get bored / tired and will do its full analysis regardless.
In the end, AIs will mostly be doing this job. It isn't a fair contest because an AI can be trained on billions of images, and every correct diagnosis, whereas a human has human limits.
reddit
AI Bias
1569427978.0
♥ 2
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | none |
| Reasoning | consequentialist |
| Policy | unclear |
| Emotion | mixed |
| Coded at | 2026-04-25T08:13:13.233606 |
Raw LLM Response
[
{"id":"rdc_ewfavla","responsibility":"company","reasoning":"unclear","policy":"unclear","emotion":"fear"},
{"id":"rdc_exfcv8r","responsibility":"unclear","reasoning":"unclear","policy":"unclear","emotion":"indifference"},
{"id":"rdc_f1e8svo","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"resignation"},
{"id":"rdc_f1eb99q","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"approval"},
{"id":"rdc_f1eqdrv","responsibility":"none","reasoning":"consequentialist","policy":"unclear","emotion":"mixed"}
]