Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
Media is happy to put the working class out of a job when it comes to automation…
ytc_UgxBUyzXt…
G
Photography, traditional art, digital art, and AI art are all different things.
…
ytr_UgxRqve2i…
G
I really appreciate the insight of this podcast and the journalist well explaine…
ytc_UgzL4JXFn…
G
Thank you, hearing from Geoffrey Hinton about the dangers of AI was very informa…
ytc_Ugzq4rxqQ…
G
How cool a robot that ruins the book i am paying for. I dont want a book with cr…
ytc_UgzKFpl4h…
G
Most of our military drones use a human control system. If it became fully autom…
ytc_UgyPBjLVe…
G
Honestly, this data is fed into the Database, and now AI will train to be better…
ytc_UgwnbIMm-…
G
And they should have to explain exactly how their programs are working and their…
ytc_Ugzebr6uL…
Comment
I can’t think of a more ill-advised move than this, sorry.
I’m an attorney working primarily in appeals, which is heavy on the legal writing. I’ve messed around with ChatGPT a few times out of curiosity. Mind you my writing is a lot more complicated than a standard contract, but the principle still applies.
ChatGPT sucks for actual legal purposes. It’s just not at that level yet. It was making up cases in support of my arguments, it was creating and applying completely fictional, entirely indefensible rules and legal tests when it couldn’t find an actual answer.
Not to mention we are a self-regulating profession working within courts and agencies that often don’t even use electronic filing, and run by people who have a deep and abiding disdain for new technology. It’s not a matter of our system not being in the 21st century- in many cases we haven’t even reached the 90’s. All it takes is one asshole judge who still lives in fear of the color television to ruin your life in some cases, tbh.
Get a lawyer. Review it with said lawyer. Fuck, get a paralegal to review it. Anything other than relying on ChatGPT. You’re going to end up fucking yourself over.
reddit
AI Responsibility
1684365275.0
♥ 392
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | none |
| Reasoning | consequentialist |
| Policy | none |
| Emotion | unclear |
| Coded at | 2026-04-25T08:33:43.502452 |
Raw LLM Response
[
{"id":"rdc_jkk8tz5","responsibility":"none","reasoning":"deontological","policy":"none","emotion":"indifference"},
{"id":"rdc_jkl4knf","responsibility":"user","reasoning":"deontological","policy":"none","emotion":"outrage"},
{"id":"rdc_jklk2x7","responsibility":"user","reasoning":"virtue","policy":"none","emotion":"outrage"},
{"id":"rdc_jkkdbon","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"disapproval"},
{"id":"rdc_jkkhh87","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"approval"}
]