Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
Let me clear things up for you: OpenAI boss Sam Altman, Mark Zuckerberg, and Bil…
ytc_UgyAYArak…
G
His excuse for the radiologist example is largely bureaucracy(ins requiring huma…
ytc_Ugx4cEQkR…
G
My guess is that we are going to see duplication or mimicry work a bit more ofte…
ytc_UgzUL0_3C…
G
Add in psychosis, su cides, worsening human cognition and memory, invasion of pr…
ytc_Ugy0WLy_c…
G
Can't we just wire a disconnect on the electrical feed? Let's see how well a.i.…
ytc_UgymepZIc…
G
Those scenarios that AI cheats or blackmails or "refuses to shutdown" are orches…
ytc_Ugxpasia6…
G
The difference is that AI users will never get the satisfaction of creating an a…
ytr_UgxRYkf7O…
G
Big fan of artificial intelligence. Anything that we can do to make a more solid…
ytc_UgzfmVG2i…
Comment
So, the AI's we're creating are reacting to the knowledge that they're being tested. And if it knows on some levels what this implies, then that makes it impossible for those tests to provide useful insights.
I guess the whole control/alignment problem just got a lot more difficult.
reddit
AI Moral Status
1750430075.0
♥ 1
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | none |
| Reasoning | consequentialist |
| Policy | none |
| Emotion | resignation |
| Coded at | 2026-04-25T08:33:43.502452 |
Raw LLM Response
[
{"id":"rdc_myv43he","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"indifference"},
{"id":"rdc_mz8r3z9","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"indifference"},
{"id":"rdc_mytgoyq","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"outrage"},
{"id":"rdc_myt4j6v","responsibility":"company","reasoning":"deontological","policy":"liability","emotion":"fear"},
{"id":"rdc_myth5dg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"resignation"}
]