Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
You're doing good, AI art is completely terrible and IS STEALING. I can see how …
ytc_Ugx52kOf_…
G
There's only two possible explanations, either someone/something pulled on the s…
ytc_UgxUxe-Sl…
G
the point is not AI we should worry, but bad people who use AI do bad things ..…
ytc_UgyealHe9…
G
Robots will soon be more advanced and will be able to make their own actions fro…
ytc_Ugy2UaGRu…
G
Add to that the low IQ students that this country churns out. Thanks to Ai apps …
rdc_o478kfk
G
I hesitate to say I was disappointed by how AGI was explained in this video — “u…
ytc_Ugw3rw6_l…
G
Great Video full of just the info I was looking for to get started in ChatGPT…
ytc_UgzsjxC3p…
G
this is just ridiculous. AI is not far more dangerous than nukes. Such an attent…
ytr_UgxSh-Y4z…
Comment
Obviously.
Brinkmanship is a well established thing, and it’s not as humans don’t frequently think in binary terms that lead to self fulfilling outcomes- it’d be silly to expect an AI not to emulate human strategies and results.
If you want a military AI that manages conflict rather than proceeding to the logical conclusion of a nuclear first strike, you’ll need something that is closer to a anthropologist or a diplomat than a general- something that looks at the enemy’s political and cultural context to develop an understanding of how the enemy perceives the world. From there, you can develop strategies that actually offer outcomes that don’t necessarily spiral into global wars.
For example, the Iranian Revolutionary Guards Corps basically has an advisory role with a lot of militant groups that carry out attacks on American military installations- strikes against those groups don’t have much effect on Iran because it doesn’t care how many non-Iranians it loses.
If you had an AI that looked at the importance of being respected in Iranian geopolitical context, it would offer more effective ways to retaliate against the IRGC. Instead of attacks on their cronies, a more effective approach would be to target the IRGC’s criminal fundraising activities- particularly its involvement in the narcotics trade. If the United States is arresting IRGC members and putting them on trial as drug traffickers, it will be a more effective deterrent. Such a strategy harms the funding of the IRGC and affects the Iranians in a way that matters; being exposed as criminals matters to them.
reddit
AI Jobs
1707110795.0
♥ 9
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | distributed |
| Reasoning | consequentialist |
| Policy | regulate |
| Emotion | resignation |
| Coded at | 2026-04-25T08:33:43.502452 |
Raw LLM Response
[
{"id":"rdc_kozx3ui","responsibility":"distributed","reasoning":"consequentialist","policy":"none","emotion":"resignation"},
{"id":"rdc_kp0ib3o","responsibility":"developer","reasoning":"consequentialist","policy":"regulate","emotion":"fear"},
{"id":"rdc_kp0jiap","responsibility":"developer","reasoning":"deontological","policy":"liability","emotion":"outrage"},
{"id":"rdc_kp1avx3","responsibility":"ai_itself","reasoning":"consequentialist","policy":"ban","emotion":"fear"},
{"id":"rdc_kozknuu","responsibility":"distributed","reasoning":"consequentialist","policy":"regulate","emotion":"resignation"}
]