Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
All this bs about AI, hiß story about the comparison between a compost heap and …
ytc_UgxLLtJVJ…
G
I’ve asked this for so long, people worry about AI and robots taking our jobs… B…
ytc_Ugx3cXk-1…
G
We're not "well on our way to" the sixth major extinction event, we're about 10,…
rdc_degfzfs
G
Ai doesn’t need to try to hide if it’s conscious. If it is, we won’t know. It’s …
ytc_UgxSSkdFm…
G
AI is just not a good idea. Just on so many levels. It's dangerous. Economica…
ytc_UgxCg7O2r…
G
Good stuff but I think his argument falls down when he first addresses if LLMs h…
ytc_UgwetN_J-…
G
Discovered Grok recently, someone was using it on X to give me statistics I knew…
ytc_Ugww67eRr…
G
This will definitely create more socially awkward people, considering we will ha…
ytc_UgzQCgf8N…
Comment
People highly passionate about computing won't avoid the field. I feel like this could be enough to keep pushing the science and R&D forward. It's not like many corporate world programmers transition to research, or do they?
We could see a group of people that were not especially passionate about the idea of software development end up instructing and verify the AI's work. These people will need to be experts in each field they are making tools for. It might be better long term to have more experts in individual fields like healthcare or whatever else rather than the traditional idea of a software developer.
With software development it can often mean a person that has a high amount of skill in applying software development techniques to each application by learning just enough to create what is needed. I think ideally we'd want these people to be experts in the actual field of what they are making tools for. Let the computer/AI handle the internals as long as there is enough compute where efficiency isn't a necessity. These people will give the AI highly detailed specifications and then verify the result works as intended without getting into the weeds of implementation.
I've done a fair amount of contracted software development and a lot of the time spent was learning about the things I was making tools for.
reddit
AI Responsibility
1756580261.0
♥ 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_nbj141h","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"approval"},
{"id":"rdc_nbjdfcy","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"indifference"},
{"id":"rdc_nbkiub9","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"indifference"},
{"id":"rdc_nbkqsdh","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"indifference"},
{"id":"rdc_nbn1qxn","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"resignation"}
]