Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
The problem with the test is that LLM AI, does not actually know what the object…
ytc_UgzTZ8p1l…
G
Even more fundamentally, this is an exogenous, rather than endogenous, shock, me…
rdc_gkqnd9t
G
Every AI agent that uses work that derives from a Human should have to pay a roy…
ytc_UgySIUiwH…
G
The real problem is that while it's super useful for any game developer who want…
ytc_UgwLaOnsC…
G
AI "artists" when their string of letters doesn't make an amazing work of art an…
ytc_UgzGuzrui…
G
I honestly just think it’s openAI dumbing down for power users. You know the one…
rdc_msa18gf
G
It definetely is stealing, these drawings put into datasets are purposeful input…
ytc_UgxncKLRP…
G
Wouldn't it be more efficient to automate trains before you automate trucks? Sin…
ytc_UgzllkrKj…
Comment
>Now when we start making AI powered automations
I’ve been in the Enterprise Automation space my entire career (almost 10 years)…and have spent past couple years integrating AI into our solutions.
What you’re talking about already exists, it’s just not as “sexy” as you would think.
A lot of “AI powered automations” are simply using AI for processing of unstructured data as context/validation for more traditional automation solutions that are programmatic.
Within the next year or two, we will continuously have more and more AI create these programmatic solutions “on the fly” as the unstructured data is synthesized.
They’re calling it “dynamic automation” in the VC circles currently, all it’s really doing is taking more and more load off the human to do simplistic yet subjective validations.
I’ve always said that Automation/AI RARELY automates 100% of ANYTHING…in fact, I advise not to even try as after a certain threshold you get diminishing returns from an enterprise perspective.
What you use AI for is to AUGMENT processes and tasks, that augmentation requires less human involvement overall once implemented, sometimes up to a degree of 95% or more (where it’s considered “straight through processing” or STP) but that is honestly rare.
Now the problem isn’t what AI will do, it’s already doing it and it’s only accelerating, the problem will be is there enough NEW WORK to be done by humans to satisfy that delta created by Automation/AI.
Traditional principles say that yes, just like how agriculture was cornerstone of our jobs a century ago was the majority, whereas now it’s the minority.
However, we’ve never had a technology that can theoretically “self-improve” (we still don’t, at least…not in any scalable way) so the past may not reflect the present.
reddit
AI Jobs
1700158086.0
♥ 1
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_k9j50lr","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"},
{"id":"rdc_k9hyn9j","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"approval"},
{"id":"rdc_k9hk3ea","responsibility":"none","reasoning":"contractualist","policy":"unclear","emotion":"outrage"},
{"id":"rdc_k9lvf2d","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"},
{"id":"rdc_k9kazjg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"approval"}
]