Raw LLM Responses

Inspect the exact model output for any coded comment.

Comment
I don't disagree that GPT4 can be a useful tool *for* a writer, but keep in mind that to properly prompt the model (prompt tinkering), you (the user) already need to be clever and know what you want. I think it's a multiplicative effect. Someone who knows nothing about writing can easily make the software do "something" but 0 skill × GPT skill is still zero. Like, check out programming forums/exchanges. Ever since GPT has been released, there's been a flood of absolute know-nothings who've forced the model to write code for them but they have absolutely no idea where to "put" the code or what to do with it. Right now, it's very much a "garbage in, garbage out" process and you need to be "smart" in the first place to use it.
reddit AI Jobs 1683178221.0 ♥ 8
Coding Result
DimensionValue
Responsibilityuser
Reasoningconsequentialist
Policynone
Emotionapproval
Coded at2026-04-25T08:33:43.502452
Raw LLM Response
[{"id":"rdc_jir9725","responsibility":"company","reasoning":"consequentialist","policy":"none","emotion":"resignation"},{"id":"rdc_jiqpgw5","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"indifference"},{"id":"rdc_jispd2j","responsibility":"user","reasoning":"consequentialist","policy":"none","emotion":"approval"},{"id":"rdc_jirmeb3","responsibility":"company","reasoning":"virtue","policy":"none","emotion":"outrage"},{"id":"rdc_jiqflxo","responsibility":"company","reasoning":"consequentialist","policy":"none","emotion":"outrage"}]